FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003
COMPLETE SPECIFICATION (See Section 10 and Rule 13)
Title of invention:
OPTIMAL CAREER PATH RECOMMENDATION USING ASPIRATION OF USERS
Applicant
Tata Consultancy Services Limited A company Incorporated in India under the Companies Act, 1956
Having address:
Nirmal Building, 9th floor,
Nariman point, Mumbai 400021,
Maharashtra, India
Preamble to the description
The following specification particularly describes the invention and the manner in which it is to be performed.

TECHNICAL FIELD [001] The disclosure herein generally relates to career path recommendations, and, more particularly, to optimal career path recommendation using aspiration of users.
BACKGROUND [002] Employees working in several organizations aspire to become better with development of skill sets both personally and professionally. Career development is an important aspect for employees’ growth and satisfaction within an organization. Career related satisfaction results in high productivity. Manual mentoring of employees for career development suffers from subjectivity, bias, less-attention and incompleteness. Further due to lack in active career mentoring/ recommendation system these employees become passive and eventually they give up on new career upgrade. There are career advisory systems available which only provide “what to do” but they do not give detailed timeline when and how to start and end a specific career upgrade activity. More specifically, these conventional career advisory systems do not offer active career guidance and do not provide personalized recommendation to the employees.
SUMMARY [003] Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems. For example, in one aspect, there is provided a processor implemented method for career path recommendation using aspiration of user. The method comprises: obtaining, via one or more hardware processors, information pertaining to a career path and an aspiration information specific to a user associated with an entity, wherein the career path comprises one or more past events associated with the user; generating and recommending, via the one or more hardware processors, at least one optimal career path with a time-bound action plan using the obtained information, wherein the step of generating and recommending the at least one optimal career path with

the time-bound action plan comprises: obtaining, via a deep learning neural network executed by the one or more hardware processors, one or more career affecting events pertaining to the user and one or more candidate mentors associated with the entity; learning, via the deep learning neural network executed by the one or more hardware processors, an embedding of each of the one or more career affecting events for a plurality of users associated with the entity, wherein the learnt embedding is indicative of an embedded mathematical representation of relationship among each of the one or more career affecting events; performing in the embedded mathematical representation, using a career edit distance technique executed by the one or more hardware processors, a first comparison of (i) the career path of the user with (ii) a past career path of the one or more candidate mentors to obtain a first personalized matching career path recommendation data, wherein the first personalized matching career path recommendation data comprises an optimally matched career path of the one or more candidate mentors based on the career path of the user; performing in the embedded mathematical representation, using a cosine similarity technique executed by the one or more hardware processors, a second comparison of (i) the aspirational information of the user with (ii) the past career path of the one or more candidate mentors to obtain a second personalized matching career path recommendation data; and generating, via the one or more hardware processors, a list of optimal career paths of the one or more candidate mentors based on the first personalized matching career path recommendation data and the second personalized matching career path recommendation data, wherein the list of optimal career paths of the one or more candidate mentors is indicative of the at least one optimal career path with the time-bound action plan for recommendation to the user.
[004] In an embodiment, the at least one optimal career path comprises at least one of (i) a sequence of one or more roles, (ii) a sequence of one or more projects, (iii) at least one training course, and (iv) at least one certification, and wherein each of the one or more role and the one or more projects comprise a recommended career timeline.

[005] In an embodiment, the one or more career affecting events comprise one or more projects being allocated to the user, one or more roles being assigned to the user, one or more training courses attended or to be attended by the user, and associated one or more certifications thereof.
[006] In an embodiment, the step of generating and recommending, via the one or more hardware processors, at least one optimal career path, is based on a domain knowledge driven technique, comprising: obtaining, via the one or more hardware processors, the one or more career affecting events pertaining to the user and the one or more candidate mentors associated with the entity; performing based on the one or more career affecting events, via the one or more hardware processors, a first comparison of (i) a career path of the user with (ii) a past career path of the one or more candidate mentors to obtain a first personalized matching career path recommendation data, wherein the first personalized matching career path recommendation data comprises information on (i) percentage of a project career path of the user matching a past project career path comprised in the past career path of the one or more candidate mentors, (ii) percentage of a role career path of the user matching a past role career path comprised in the past career path of the one or more candidate mentors; (iii) percentage of training attended or training in progress, and certifications received by the user matching training courses and certifications comprised in the past career path of the one or more candidate mentors; performing, via the one or more hardware processors, a second comparison of (i) the aspirational information of the user with (ii) the past career path of the one or more candidate mentors to obtain a second personalized matching career path recommendation data, wherein the second personalized matching career path recommendation data comprises information on (a) percentage of the skill and domain aspiration information of the user matching project career path comprised in the past career path of the one or more candidate mentors, (b) percentage of role aspiration information of the user matching with project career path or role career path comprised in the past career path of the one or more candidate mentors, (c) percentage of skills aspiration information of the user matching a training career path comprised in the past career path of the one or more candidate mentors, and

(d) percentage of skills aspiration information of the user matching a certification career path comprised in the past career path of the one or more candidate mentors; and generating a list of optimal career paths of the one or more candidate mentors based on the first personalized matching career path recommendation data and the second personalized matching career path recommendation data, wherein the list of optimal career paths of the one or more candidate mentors is indicative of the at least one optimal career path with the time-bound action plan for recommendation to the user.
[007] In another aspect, there is provided a system for career path recommendation using aspiration of user. The system comprises: a memory storing instructions; one or more communication interfaces; and one or more hardware processors coupled to the memory via the one or more communication interfaces, wherein the one or more hardware processors are configured by the instructions to: obtain information pertaining to a career path and an aspiration information specific to a user associated with an entity, wherein the career path comprises one or more past events associated with the user; generate and recommend at least one optimal career path with a time-bound action plan using the obtained information, wherein the at least one optimal career path with the time-bound action plan is generated and recommended by: obtaining, via a deep learning neural network executed by the one or more hardware processors, one or more career affecting events pertaining to the user and one or more candidate mentors associated with the entity; learning, via the deep learning neural network executed by the one or more hardware processors, an embedding of each of the one or more career affecting events for a plurality of users associated with the entity, wherein the learnt embedding is indicative of an embedded mathematical representation of relationship among each of the one or more career affecting events; performing in the embedded mathematical representation, using a career edit distance technique executed by the one or more hardware processors, a first comparison of (i) the career path of the user with (ii) a past career path of the one or more candidate mentors to obtain a first personalized matching career path recommendation data, wherein the first personalized matching career path recommendation data comprises an optimally

matched career path of the one or more candidate mentors based on the career path of the user; performing in the embedded mathematical representation, using a cosine similarity technique executed by the one or more hardware processors, a second comparison of (i) the aspirational information of the user with (ii) the past career path of the one or more candidate mentors to obtain a second personalized matching career path recommendation data; and generating, via the one or more hardware processors, a list of optimal career paths of the one or more candidate mentors based on the first personalized matching career path recommendation data and the second personalized matching career path recommendation data, wherein the list of optimal career paths of the one or more candidate mentors is indicative of the at least one optimal career path with the time-bound action plan for recommendation to the user.
[008] In an embodiment, the at least one optimal career path comprises at least one of (i) a sequence of one or more roles, (ii) a sequence of one or more projects, (iii) at least one training course, and (iv) at least one certification, and wherein each of the one or more role and the one or more projects comprise a recommended career timeline.
[009] In an embodiment, the one or more career affecting events comprise one or more projects being allocated to the user, one or more roles being assigned to the user, one or more training courses attended or to be attended by the user, and associated one or more certifications thereof.
[010] In an embodiment, the at least one optimal career path is generated and recommended using a domain knowledge driven technique by: obtaining, via the one or more hardware processors, the one or more career affecting events pertaining to the user and the one or more candidate mentors associated with the entity; performing based on the one or more career affecting events, via the one or more hardware processors, a first comparison of (i) a career path of the user with (ii) a past career path of the one or more candidate mentors to obtain a first personalized matching career path recommendation data, wherein the first personalized matching career path recommendation data comprises information on (i) percentage of a project career path of the user matching a past project career path

comprised in the past career path of the one or more candidate mentors, (ii) percentage of a role career path of the user matching a past role career path comprised in the past career path of the one or more candidate mentors; (iii) percentage of training attended or training in progress, and certifications received by the user matching training courses and certifications comprised in the past career path of the one or more candidate mentors; and performing, via the one or more hardware processors, a second comparison of (i) the aspirational information of the user with (i) the past career path of the one or more candidate mentors to obtain a second personalized matching career path recommendation data, wherein the second personalized matching career path recommendation data comprises information on (a) percentage of the skill and domain aspiration information of the user matching project career path comprised in the past career path of the one or more candidate mentors, (b) percentage of role aspiration information of the user matching with project career path or role career path comprised in the past career path of the one or more candidate mentors, (c) percentage of skills aspiration information of the user matching a training career path comprised in the past career path of the one or more candidate mentors, and (d) percentage of skills aspiration information of the user matching a certification career path comprised in the past career path of the one or more candidate mentors; and generating a list of optimal career paths of the one or more candidate mentors based on the first personalized matching career path recommendation data and the second personalized matching career path recommendation data, wherein the list of optimal career paths of the one or more candidate mentors is indicative of the at least one optimal career path with the time-bound action plan for recommendation to the user.
[011] In yet another aspect, there are provided there are provided one or more non-transitory machine-readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors cause career path recommendation using aspiration of user by obtaining information pertaining to a career path and an aspiration information specific to a user associated with an entity, wherein the career path comprises one or more past events associated with the user; generating and recommending, via the one or more

hardware processors, at least one optimal career path with a time-bound action plan using the obtained information, wherein the step of generating and recommending the at least one optimal career path with the time-bound action plan comprises: obtaining, via a deep learning neural network executed by the one or more hardware processors, one or more career affecting events pertaining to the user and one or more candidate mentors associated with the entity; learning, via the deep learning neural network executed by the one or more hardware processors, an embedding of each of the one or more career affecting events for a plurality of users associated with the entity, wherein the learnt embedding is indicative of an embedded mathematical representation of relationship among each of the one or more career affecting events; performing in the embedded mathematical representation, using a career edit distance technique executed by the one or more hardware processors, a first comparison of (i) the career path of the user with (ii) a past career path of the one or more candidate mentors to obtain a first personalized matching career path recommendation data, wherein the first personalized matching career path recommendation data comprises an optimally matched career path of the one or more candidate mentors based on the career path of the user; performing in the embedded mathematical representation, using a cosine similarity technique executed by the one or more hardware processors, a second comparison of (i) the aspirational information of the user with (ii) the past career path of the one or more candidate mentors to obtain a second personalized matching career path recommendation data; and generating, via the one or more hardware processors, a list of optimal career paths of the one or more candidate mentors based on the first personalized matching career path recommendation data and the second personalized matching career path recommendation data, wherein the list of optimal career paths of the one or more candidate mentors is indicative of the at least one optimal career path with the time-bound action plan for recommendation to the user. [012] In an embodiment, the at least one optimal career path comprises at least one of (i) a sequence of one or more roles, (ii) a sequence of one or more projects, (iii) at least one training course, and (iv) at least one certification, and

wherein each of the one or more role and the one or more projects comprise a recommended career timeline.
[013] In an embodiment, the one or more career affecting events comprise one or more projects being allocated to the user, one or more roles being assigned to the user, one or more training courses attended or to be attended by the user, and associated one or more certifications thereof.
[014] In an embodiment, the step of generating and recommending, via the one or more hardware processors, at least one optimal career path, is based on a domain knowledge driven technique, comprising: obtaining, via the one or more hardware processors, the one or more career affecting events pertaining to the user and the one or more candidate mentors associated with the entity; performing based on the one or more career affecting events, via the one or more hardware processors, a first comparison of (i) a career path of the user with (ii) a past career path of the one or more candidate mentors to obtain a first personalized matching career path recommendation data, wherein the first personalized matching career path recommendation data comprises information on (i) percentage of a project career path of the user matching a past project career path comprised in the past career path of the one or more candidate mentors, (ii) percentage of a role career path of the user matching a past role career path comprised in the past career path of the one or more candidate mentors; (iii) percentage of training attended or training in progress, and certifications received by the user matching training courses and certifications comprised in the past career path of the one or more candidate mentors; performing, via the one or more hardware processors, a second comparison of (i) the aspirational information of the user with (ii) the past career path of the one or more candidate mentors to obtain a second personalized matching career path recommendation data, wherein the second personalized matching career path recommendation data comprises information on (a) percentage of the skill and domain aspiration information of the user matching project career path comprised in the past career path of the one or more candidate mentors, (b) percentage of role aspiration information of the user matching with project career path or role career path comprised in the past career path of the one or more candidate mentors, (c)

percentage of skills aspiration information of the user matching a training career path comprised in the past career path of the one or more candidate mentors, and (d) percentage of skills aspiration information of the user matching a certification career path comprised in the past career path of the one or more candidate mentors; and generating a list of optimal career paths of the one or more candidate mentors based on the first personalized matching career path recommendation data and the second personalized matching career path recommendation data, wherein the list of optimal career paths of the one or more candidate mentors is indicative of the at least one optimal career path with the time-bound action plan for recommendation to the user.
[015] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
BRIEF DESCRIPTION OF THE DRAWINGS
[016] The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles:
[017] FIG. 1 depicts an exemplary block diagram of a system for optimal career path recommendation using aspiration of users, in accordance with an embodiment of the present disclosure.
[018] FIG. 2 depicts an exemplary flow chart for optimal career path recommendation using aspiration of users using the system of FIG. 1, in accordance with an embodiment of the present disclosure.
[019] FIG. 3 illustrates an exemplary deep learning neural network (e.g., a long short-term memory neural network (LSTM-NN)) for similarity measure and generation and recommendation of the optimal career path to the user thereof, in accordance with an embodiment of the present disclosure.
[020] FIG. 4 depicts an example for achievement duration in accordance with an embodiment of the present disclosure.

DETAILED DESCRIPTION OF EMBODIMENTS [021] Exemplary embodiments are described with reference to the accompanying drawings. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the scope of the disclosed embodiments. It is intended that the following detailed description be considered as exemplary only, with the true scope being indicated by the following claims.
[022] Across industries, technologies as well as products and services offerings have been changing rapidly, leading to changes in work profiles, roles, tasks, locations, responsibilities. With changing on-the-job work profiles and continuous skilling and re-skilling, employees now have a wider choice of career options. Human Resource (HR) systems nowadays collect data on all aspects of employees’ work history, so that organizations are now in a unique position to offer career guidance to employees and help them meet their career aspirations. Being on a career path not matching with own aspirations is one of the major causes of employee dissatisfaction, and such timely and accurate career advice will lead to engaged and satisfied employees, and reduced attrition.
[023] Currently almost all career advice comes from mentors, seniors, friends and HR executives. Alternatively, an employee identifies a senior employee as a role model, and “follows” his/her trajectory as much as possible. Such approaches suffer from subjectivity and come without much idea of how likely the employee can follow recommendations or to achieve own aspirations. Embodiments of the present disclosure provide systems and methods that implement data-driven technique(s) that use the career paths of all employees, present and past, to generate career advice for an employee. Basically, in the present disclosure, past data of career paths is mined or analyzed to identify similar but senior employees, who once were at a career stage similar to a querying employee, say ‘q’ and who have later attained the aspirations specified by ‘q’. The career

advice is distilled from their career paths and contains recommended trainings, roles, projects, and the like with suggested durations. Special emphasis is made by the present disclosure on the explicitly stated career aspirations of an employee. The present disclosure implements a domain-knowledge that captures insights from mentoring, and a deep learning method(s) to discover embeddings for career elements (e.g., roles, skills, and the like) and then uses them.
[024] More specifically, embodiments of the present disclosure implement the data-driven technique(s) to guide an individual employee to achieve her aspiration within the organization. Present disclosure also provides personalized recommendation to ensure smooth and realistic career transition/upgrade. A ‘model career path’ is generated along with a timeline by system of the present disclosure, which is a refined career path of an existing or past employee in the same organization. To an employee who is looking for career upgrade, the system of the present disclosure recommends a ‘model career path’ whose trajectory best matches with querying employee’s career path and his/her stated aspiration information. Further, systems and methods of the present disclosure define (i) ‘career similarity’ between query employee’s career path and the past career trajectory of a candidate model career path, and (ii) ‘aspiration similarity’ between aspirations of querying employee to recent career path of candidate model’s career path, wherein both measures (e.g., the career similarity and the aspiration similarity) are utilized to generate and recommend an optimal career path for an employee looking for recommendation. The generated optimal career path being recommended includes a timeline of career activities in terms of project, skills, roles, trainings and certifications which are to be performed timely by a specific employee to get aspired upgrade in his/her career.
[025] Referring now to the drawings, and more particularly to FIGS. 1 through 4, where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments and these embodiments are described in the context of the following exemplary system and/or method.

[026] FIG. 1 depicts an exemplary block diagram of a system 100 for optimal career path recommendation using aspiration of users, in accordance with an embodiment of the present disclosure. The system 100 may also be referred as ‘analysis system’ and may be interchangeably used hereinafter. In an embodiment, the system 100 includes one or more hardware processors 104, communication interface device(s) or input/output (I/O) interface(s) 106 (also referred as interface(s)), and one or more data storage devices or memory 102 operatively coupled to the one or more hardware processors 104. The one or more processors 104 may be one or more software processing components and/or hardware processors. In an embodiment, the hardware processors can be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the processor(s) is configured to fetch and execute computer-readable instructions stored in the memory. In an embodiment, the system 100 can be implemented in a variety of computing systems, such as laptop computers, notebooks, hand-held devices, workstations, mainframe computers, servers, a network cloud and the like.
[027] The I/O interface device(s) 106 can include a variety of software and hardware interfaces, for example, a web interface, a graphical user interface, and the like and can facilitate multiple communications within a wide variety of networks N/W and protocol types, including wired networks, for example, LAN, cable, etc., and wireless networks, such as WLAN, cellular, or satellite. In an embodiment, the I/O interface device(s) can include one or more ports for connecting a number of devices to one another or to another server.
[028] The memory 102 may include any computer-readable medium known in the art including, for example, volatile memory, such as static random access memory (SRAM) and dynamic random access memory (DRAM), and/or non-volatile memory, such as read only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes. In an embodiment, a database 108 is comprised in the memory 102, wherein the database

108 comprises information, for example, a career path and an aspiration information specific to a user associated with an entity, wherein the career path comprises one or more past events associated with the user. The information stored in the database 108 may further comprise career affecting events pertaining to the user and one or more candidate mentors associated with the entity wherein embedding of each of the one or more career affecting events for a plurality of users associated with the entity are learnt and stored in the database 108. The one or more career affecting events comprise projects being allocated to the user, one or more roles being assigned to the user, one or more training courses attended or to be attended by the user and associated one or more certifications thereof. The memory 102 further comprises comparison results (e.g., a first comparison of (i) the career path of the user with (ii) a past career path of the one or more candidate mentors, and a second comparison of (i) the aspirational information of the user with (ii) the past career path of the one or more candidate mentors). The memory 102 further comprises at least one optimal career path being recommended to the user, wherein the optimal career path comprises at least one of (i) one or more roles sequence, (ii) one or more projects sequence, (iii) at least one training course, and (iv) at least one certification, and wherein each of the one or more role sequences and the one or more project sequence comprise a recommended career timeline.
[029] In an embodiment, the memory 102 may store (or stores) one of more techniques. For instance, career edit distance technique(s), cosine similarity, and the like. The memory further comprises a deep learning neural network (e.g., a long short-term memory LSTM neural network), a domain knowledge driven technique and the like which when get executed perform one or more methodologies described hereinafter. The memory 102 further comprises (or may further comprise) information pertaining to input(s)/output(s) of each step performed by the systems and methods of the present disclosure. In other words, input(s) fed at each step and output(s) generated at each step are comprised in the memory 102 and can be utilized in further processing and analysis.
[030] FIG. 2, with reference to FIG. 1, depicts an exemplary flow chart for optimal career path recommendation using aspiration of users using the system 100

of FIG. 1, in accordance with an embodiment of the present disclosure. In an embodiment, the system(s) 100 comprises one or more data storage devices or the memory 102 operatively coupled to the one or more hardware processors 104 and is configured to store instructions for execution of steps of the method by the one or more processors 104. The steps of the method of the present disclosure will now be explained with reference to components of the system 100 of FIG. 1, the flow diagram as depicted in FIG. 2 and FIGS. 3 through 4.
[031] Prior to discussing the steps being carried out by the present disclosure, the present disclosure formulates a problem for career path recommendation by way of below illustrative example:
[032] In an information technology (IT) industry, a skill typically denotes programming languages (e.g., Python), technology platforms (e.g., Linux, Oracle), software tools (e.g., GNU Linear Programming Kit (GLPK), JTest - automated Java software testing and static analysis), generic technical skills (e.g., software testing, Graphical User Interface (GUI) design), etc. Let SK denote the set of all skills. It is assumed that a grouping of skills into skill clusters is given, where each skill cluster contains “closely related” skills, e.g., skill cluster JAVA may contain skills Core Java, J2EE, Spring, Hibernate. etc. Let SC denote the set of all skill clusters, where for every element C ∈SC, C⊆SK. Skill clusters need not be disjoint i.e., the same skill may belong to two or more skill clusters. A role denotes a specific set of tasks, responsibilities and deliverables expected from an employee in a project; examples of roles: DEVELOPER, PROJECT LEADER, TESTENGINEER, DBA, REQANALYST, etc. Let RO denote the set of all possible roles. Let DOM denote a set of industry segments (called domains), such as Banking, Insurance, Retail, Telecom, etc.
[033] Let ET denote the set of types of career affecting events; wherein in the present disclosure, ET = {project,role_set, cert}. Here, project_alloc denotes the allocation of the employee to a project, role_set denotes the assignment of a role to an employee, training denotes the employee attending a training program in a particular skill, and cert denotes the employee passing an external examination for a skill. Since focus is on career paths within an

organization, present disclosure considers only those event types that occur within an organization. Thus, change of job is not considered as an event type.
[034] Each event type has associated with it a tuple of event arguments, which provide additional information, as follows. For example, an event of type project_alloc has attributes: duration of allocation, an industry domain and set of skill clusters denoting the skills used in that project as described below: project_alloc: N × DOM × 2SC role_set: RO × N training: SC×SK �role_set:SC×SK|
[035] An event mention (or event) identifies the occurrence of a specific type of career affecting event for a specific employee. Event mentions are durative (have start and end dates) and stative (indicate a condition or state that holds true for that employee from the associated start and end dates). Formally, an event (mention) is a tuple (T,atg,start_date,end_date), where T ∈ ET is the event type, arg is the tuple of event arguments, and start_date, end_date are dates (inclusive) on which this event began and ended. For example, (Sep_2017, (183, Banking, {JAVA, ORACLE}), 01 - Apr - 2017,30 -Sep - 2017) is an event where an employee joined a project in the Banking domain on 01 -Apr - 2017 and worked in this project till 30 - Sep - 2017. The project used skills in the clusters JAVA, ORACLE, and has the duration of 183 days.
[036] A career path (CP), for a specific employee ei is an ordered sequence of events σi = (c1, c2,…,cni), such that the start_date of the events are in non-decreasing temporal order i.e., cj.start_date ≤ ck.start_date for every 1≤j,k≤ni such that j < k. In general, events within a CP can overlap. For example, an employee can do a training while he/she is allocated to a project. Duration of an event is the number of days between its start_date and end_date.
[037] Some operators are defined on career paths by the present disclosure. The projection operator π(σ,T) removes all events from the givenCP except those of type T ∈ ET, and returns the resulting sequence, which is clearly a

subsequence of The prefix (suffix) operator removes all
events from the given except those whoseis on or before (after)
the given date t, and returns the resulting sequence, which is clearly a contiguous subsequence of For a given the prefix and suffix operators are inverses of
each other is the sequence concatenation
operator. for a K-partition of a if (i) each of
is a contiguous subsequence of σ and (ii) are mutually
disj oint; and (iii) When is a prefix (suffix) of σ.
Some of the are allowed to be empty subsequence
[038] Career state of an employee having at time t is her CP up to t
which is Note that last eventsmay have to be truncated to “stop” at
the given date t. Aspiration of an employee is a tuple where
is the desired role, is the desired industry domain, andis the
list of desired skill clusters the employee wants to specialize in.
[039] It is assumed by systems and methods of the present disclosure that a dataset containing CPs of all present and past employees is
available. Let tcuteoff denote the cut-off date for D i.e., the date on which the “snapshot” of the career paths was taken. Then the for all open events in
a given CP is assumed to be set to For example, suppose
and suppose an employee got allocated to a project which will continue beyond the cutoff date, then the end_date for this event to 31 -Dec-2017. This is a form of right censoring. Let denote a similarity
function that computes the similarity between two given CPs, σ1 and σ2, returning a value in [0,1]. Two CPs, σ1 and σ2 are highly similar if sim where
is a given threshold constant. Let denote a similarity
function that computes the similarity between a given and a given aspiration
returning a value in and aspiration are highly similar if
spsim where is a given threshold constant.
[040] Present disclosure discusses ways to compute the similarity functions sim and aspsim later. Let e be a query employee having

(indicating her current career state) and aspiration p. Let e′ be any other employee
whose CP σ′ is present in D. Let’s say e′ is a model employee for e if there exists
a 4-partition
is highly similar to the current career state (called current state
similarity between e and e, and denoted fsim(e, e′)). This notion of a model employee can be further generalized by considering a 5-partition of σ′ and allowing a subsequence between a2 and a3.
[041] The problem the present disclosure solves is as follows. Given an employee a2, e, her current career state σ, aspiration p, and a dataset D of CPs, given constants and a given integer k> 0, find top k model
employees e′ in D for e, ranked in terms of the score csim(e, e′) + f sim(e, e′). Essentially, a personalized, optimal, future CP (referred as an ideal CP) is identified for e such that if e follows this CP starting from her current career state, then there is high probability that she achieves her aspirations at the end of a. This ideal CP for e can be constructed by appropriately taking the best points from the CPs of the k model employees identified for e. More specifically, the above problem being solved in better understood by way of the following steps described herein.
[042] In an embodiment, at step 202 of the present disclosure, the one or more hardware processors 104 obtain information pertaining to a career path and an aspiration information specific to a user associated with an entity (e.g., an organization and the expressions entity and organization may be interchangeably used hereinafter), wherein the career path comprises one or more past events associated with the user. In an embodiment of the present disclosure, at step 204, the one or more hardware processors 104 generate and recommend at least one optimal career path with a time-bound action plan using the obtained information. In the present disclosure, the step of generating and recommending the at least one optimal career path with the time-bound action plan is performed by utilizing one or more technique(s). More specifically, in the present disclosure, the generation and recommendation of the at least one optimal career path with the time-bound

action plan is based on implementation of at least one of a deep learning neural network, and a domain knowledge driven approach.
[043] Below steps 204(a) till (204e) describe the step of generating and recommending the at least one optimal career path with the time-bound action plan using the deep learning neural network. For instance, at step 204(a) of the present disclosure, the deep learning neural network is executed by the one or more hardware processors 104 to obtain one or more career affecting events pertaining to the user and one or more candidate mentors associated with the entity. In an embodiment of the present disclosure, the one or more career affecting events comprise one or more projects being allocated to the user, one or more roles being assigned to the user, one or more training courses attended or to be attended by the user, and associated one or more certifications thereof.
[044] At step 204(b) of the present disclosure, the deep learning neural network via the one or more hardware processors 104 learns an embedding of each of the one or more career affecting events for a plurality of users associated with the entity, wherein the learnt embedding is indicative of an embedded mathematical representation of relationship among each of the one or more career affecting events.
[045] At step 204(c) of the present disclosure, the one or more hardware processors 104 execute using the career edit distance technique comprised in the memory 102 which performs a first comparison of (i) the career path of the user with (ii) a past career path of the one or more candidate mentors to obtain a first personalized matching career path recommendation data. The first comparison is performed in the embedded mathematical representation, in one embodiment of the present disclosure. In an embodiment, the first personalized matching career path recommendation data comprises an optimally matched career path of the one or more candidate mentors based on the career path of the user. In other words, the first comparison is indicative of determining a career point in candidate mentor’s career path until that point where user’s career path could be optimally matched to candidate mentor’s career path. Anything before or until the match, the career path of the candidate mentor is referred as past career path and after the match, timeline

indicates current career path of the candidate mentor where most aspiration information is present in the current career path of the candidate mentor.
[046] At step 204(d) of the present disclosure, the one or more hardware processor 104 perform in the embedded mathematical representation, a second comparison of (i) the aspirational information of the user with (ii) the past career path of the one or more candidate mentors to obtain a second personalized matching career path recommendation data. The second comparison is performed by using the cosine similarity technique comprised in the memory 102, in one example embodiment. The second personalized matching career path recommendation data comprises information based on user preferences.
[047] At step 204(e) of the present disclosure, the one or more hardware processors 104 generate a list of optimal career paths of the one or more candidate mentors based on the first personalized matching career path recommendation data and the second personalized matching career path recommendation data, wherein the list of optimal career paths of the one or more candidate mentors is indicative of the at least one optimal career path with the time-bound action plan for recommendation to the user. More specifically, the first personalized matching career path recommendation data and the second personalized matching career path recommendation data are combined to generate the at least one optimal career path and is recommended thereof. In an embodiment of the present disclosure, the at least one optimal career path comprises at least one of (i) one or more roles sequence, (ii) one or more projects sequence, (iii) at least one training course, and (iv) at least one certification, and wherein each of the one or more role sequences and the one or more project sequence comprise a recommended career timeline. The above steps 204(a) till 204(e) are better understood by way of following description.
[048] Present disclosure and its systems (e.g., system 100) implements a deep learning technique (e.g., a long short-term memory (LSTM) neural network comprised in the memory 102) that (i) learns an embedding representation for each event mention, and computes similarity between event mentions based on their embedding; and (ii) finds a better alignment between two CPs based on the temporal

information in them and then computes the similarity. Embedding for event mentions (values) is computed by projecting all CPs for a particular event type. Such segregation of event types reduces the effects of data sparsity in multi-attribute data. The following intuitions are captured:
1. Event transitions are rarely random: If a CP contains a transition e.g., (DEVELOPER, MODULE_LEADER) then there is a good real-life reason for this transition, which implies DEVELOPER and MODULE_LEADER are related and hence have some similarity.
2. Closer events are more related: Events which are temporally nearer in a CP are more similar to each other as compared to events which are temporally farther.
3. Events having common prefix (suffix) sequences are similar: For example, if some role CPs contain and some others contain then roles b and c should have some similarity. As another example, if some roleCPs contain , and some others contain , then roles d and e should be considered somewhat similar. In contrast, if prefixes of role x has nothing in common with the prefixes of role y then x and y should be considered dissimilar.
[049] As discussed, embodiments of the present disclosure and its systems and methods implement a deep learning neural network (e.g., the LSTM neural network) for similarity measure and generation and recommendation of the optimal career path to the user thereof as depicted in FIG. 3 to learn the embedding associated with the vocabulary of a specific event type (e.g., roles). Note that these are learned based on the temporal positions of the event mentions. Available ��s sequence data (e.g., 24184 CPs) was used by the system 100 to learn the efficient embedding of the event mentions. Contiguous repeating mentions of the same role was used by a single instance of the mention (this is because the focus was on the transitions).
[050] On average each career path contains 2.4 role assignments. There were 358 distinct roles, for which the embeddings were learnt. LSTM neural network depicted in FIG. 3 was used in supervised setting to learn the embedding,

where the class label indicates valid or invalid projected C P. The reason is the supervised representation learning with negative sampling is known to be faster and more accurate. Learning embedding for training and certification names are bit more complicated than stated above. In the present disclosure, by implementing the LSTM neural network, embedding for the training and certification is learnt in two phases. In the first phase, using the Negative Sampling technique with LSTM neural network, embedding for words present in a training name are learnt. Thereafter average of each words’ embedding is computed as the embedding vector of the training name. In second phase, the embedding of each training name is refined/tuned using the training sequence present in the CPs. A similar neural network architecture is used except that it is started with initial embedding vector learned in the first phase. Given role CPs is treated as valid and the negative sampling technique was used to randomly generate invalid role sequences. In the present disclosure, the LSTM neural network architecture as depicted in FIG. 3, comprised of an embedding layer, an LSTM layer, and the last output layer with sigmoid function. This model (e.g., LSTM neural network) is trained to learn the embedding for all types of event mentions.
[051] Given role CPs were treated as valid and a negative sampling technique as known in the art was utilized to randomly generate 24184 invalid role sequences. The LSTM neural network was used and the model was trained with 10-fold cross-validation accuracy of 98.15% for the valid and invalid sequence classification. 30-dimensional learned embedding vector was extracted for each role name from the embedding layer. Similarly, the embeddings for domain names, projects and skill clusters was learnt with training accuracies of 98.1, 87 and 99.14 respectively. There were 358 distinct roles, 19 distinct domain names, and 300 distinct skill clusters.
[052] With the method as described earlier, the embedding vector for all of the event mentions of all event types are obtained. Given a projected CP σ, an embedded CP γ(σ) is the sequence CP except that each event name in σ is replaced by its embedding vector, along with the rest of the event attributes discussed above

(e.g., duration). Thus, a given CP σ has associated with four projected embedded CPs, one for each event type.
[053] Now, given two embedded projected CPs (e.g., role CPs) γ(σ1) and γ(σ2), an equivalent version of below equations. (2) to (4) and equations (6) to (8) are required. Basically, Levenshtein edit distance is computed between γ(σ1) and
γ(σ2), where the cost of editing a vector to another vector is given by the cosine distance (i.e., 1 - cosine similarity) between those vectors. The cost of editing one
duration τ1 to another Τ2 is given by a simple rule These two costs are
added as the total editing cost of one edit operation. To simply, other event attributes are disregarded.
[054] Below provided is an exemplary pseudo code for deep learning neural network based optimal career path recommendation using aspiration of users:
Se= Aspirant e’s career path; S: list of candidate career paths; Ae = Aspiration of employee e

[055] Below provided is an exemplary pseudo code for performing the first comparison of 2 career paths (e.g., career path of the user and past career path of candidate mentors) using the career edit distance/similarity technique comprised in the memory 102:
Se = Aspirant e’s career path; Si: a candidate career path Function sim careers (Se, Si): Sub_paths_e= groupByEventType(Se)

Sub_paths_i= groupByEventType(Si)
sim=0
For each pair (p, q) in (Sub_paths_e, Sub_paths_i)
dist = CareerEditSim (p,q)
sim= sim+ (1-dist) return sim/4
[056] Below provided is an exemplary pseudo code for performing the second comparison of 2 career paths using the career edit distance/similarity technique comprised in the memory 102: Ae= Aspiration of employee e Si= a candidate career path
Function computeAspriationalSim (Ae , Si): a.role = Ae.role a.domain = Ae. domain a.skill_clusters =Ae.Skill_clusters role_sim = computAspRoleSim(a.role, Si.role)
domain_sim =computeAspDomainSim(a.domain, Si. project)
skill_clusters_sim =computeAspScSim(a.skill_clusters, Si.project, Si. trainings,Si. certifications)
fin_sim = (role_sim + domain_sim + skill_clusters_sim)/3 return fin_sim

[057] Below provided is an exemplary pseudo code of the career edit distance/similarity technique comprised in the memory 102: Funciton CareersEditSim(p, q) l1= length(p) +1 l2 = length(q) +1


[058] Below description is provided for the step of generating and recommending the at least one optimal career path with the time-bound action plan using the domain knowledge driven approach. For instance, as described in the deep learning approach, the one or more career affecting events pertaining to the user and one or more candidate mentors associated with the entity are obtained. Further, the first comparison is performed between (i) a career path of the user with (ii) a past career path of the one or more candidate mentors to obtain a first personalized matching career path recommendation data. The career affecting events are used for performing the first comparison, in one example embodiment. In other words, the one or more career affecting events pertaining to the user and the one or more candidate mentors associated with the entity are obtained for performing the first comparison to obtain the first personalized matching career path recommendation data. More specifically, the first personalized matching career path recommendation data includes information on how much percentage of (i) project career path of the user matches past project career path of the one or more candidate

mentors, (ii) role career path of the user matches past role career path of the one or more candidate mentors, and (iii) training attended or being attending (or training in progress) and certifications received (or certifications in progress) by the user matches training courses and certifications of the one or more candidate mentors.
[059] More specifically, percentage of project career path of the user matching the past project career path of the one or more candidate mentors is computed by solving the below equation illustrated by way of example:

wherein is a similarity measure between at least two project
career paths, wherein is a projection of a project career path of the user, wherein is a projection of the project career path of at least one candidate mentor, wherein T1 is a total duration of wherein T2 is a total duration of
wherein ��� is an intersection of one or more domains in and
wherein m is number of events in i is an iterator, pi. d is a duration in an ith
project event of wherein n is number of events in wherein qj. d is a
duration in an jth project event of wherein C is an intersection of skills in
and , wherein C is an instance of C, wherein pi. L is a set of skill clusters associated with the ith project event of the user, and wherein qj. L is a set of skill clusters associated with the jth project event of the at least one candidate mentor.
[060] More specifically, above equation (1) enables the system 100 to compute the project career path similarity between two career paths, wherein the system 100 computes the project career path similarity by combining the matching of domain and skill along with durations. The system 100 further aggregates the overlapping duration (by taking minimum duration) between two project career

paths for all matching domain. Similarly, the system 100 aggregates the skill duration overlap of career paths. Both aggregated overlapping duration are summed up and then normalized by double of maximum of total duration between both to compute the project career path similarity.
[061] Similarly, percentage of role career path of the user matching the past role career path of the one or more candidate mentors is computed by solving the below equation illustrated by way of example:

wherein is a similarity measure between at least two role career
paths, wherein is a projection of a role career path of the user, wherein is a projection of the role career path of at least one candidate mentor, wherein T1 is
a total duration of wherein T2 is a total duration of whereinσ(r)2 is an
intersection of roles in and wherein r is an instance of R, and wherein
pi.ro is a role mention in the ith project event, qj. ro is a role mention in the jth project event.
[062] More specifically, above equation (2) computes role similarity between the two career paths. It aggregates the overlapping common roles’ duration then normalizes it by max of total duration between two role career paths.
[063] Similarly, percentage of training attended or being attending matching training courses of the one or more candidate mentors is computed by solving the below equation illustrated by way of example:

wherein is a similarity measure between at least two training
career paths, wherein is a projection of a training career path of the user, wherein is a projection of a training career path of at least one candidate

mentor, wherein C1 is a set of trainings in and wherein C2 is a set of
trainings in
[064] Above equation (3) gives the system 100 a ratio of common trainings mentioned in both career paths versus union of all trainings by both, wherein the system 100 measures the training similarity between two career paths.
[065] It is to be understood that similarly, percentage of certifications received by the user matching certifications of the one or more candidate mentors
can be computed and expressed as
[066] And the first comparison resulting in the first personalized matching career path recommendation data is obtained by taking average or combination of equations (1) to (3) along with the percentage computation of certification and is expressed by way of following exemplary equation:

Above equation (4) enables the system 100 to compute the average of sub-similarity components to compute the final similarity score.
[067] In the domain knowledge driven approach, based on the career affecting events obtained, the second comparison is performed between (i) the aspirational information of the user and (ii) the past career path of the one or more candidate mentors to obtain a second personalized matching career path recommendation data. The career affecting events are used for performing the second comparison, in one example embodiment. In other words, the one or more career affecting events pertaining to the user and the one or more candidate mentors associated with the entity are obtained for performing the second comparison to obtain the second personalized matching career path recommendation data. In an embodiment of the present disclosure, the second personalized matching career path recommendation data comprises information on how much percentage of (i) skill and domain aspiration information of the user matches project career path comprised in the past career path of the one or more candidate mentors, (ii) role

aspiration information of the user matches with project or role career path comprised in the past career path of the one or more candidate mentors, and (iii) skills aspiration information of the user matches training and certification career path comprised in the past career path of the one or more candidate mentors.
[068] More specifically, percentage of the skill and domain aspiration information of the user matching project career path comprised in the past career path of the one or more candidate mentors is determined in accordance with the following equation provided by way of example below:

wherein aspsimp(σ(p),p) is a similarity measure of aspiration of the user and the
project career path of the at least one candidate mentor, wherein is the project career path of the at least one candidate mentor and p is the aspiration of the user, wherein T is a total duration of wherein C1 is a set of skill clusters inand
C2 is a set of skill clusters in p respectively, wherein x and y are instances of C1 and C2 respectively, wherein is a similarity measure between x and y, wherein
is a set of skill clusters in the ith project event of the at least one candidate
mentor, wherein is a duration of the ith project event of, wherein
dm is a domain of the ith project event of the at least one candidate mentor, and wherein p. d is an aspiration domain of the user.
[069] More specifically, above equation (5) enables the system 100 to compute amount of user’s aspiration information similar with project career path of candidate mentor. The system 100 computes the similarity by combining both aspirational skill and domain similarity of user’s aspiration with project’s skill and domain exposer of the mentor. For aspirational skill similarity, for all combination between skills of user’s aspiration and mentor’s project skill, the system 100 computes weighted (skill similarity) and aggregated duration of the mentor’s projects. Then the system 100 normalizes it by multiplication of number of skills in user’s aspiration, mentor’s project, and total duration of the mentor’s project career

path to compute the aspirational skill similarity. Similarly, for domain similarity the system 100 aggregates overlapping domain duration of mentor’s project career path with user’s aspirational domain and then normalizes it by total duration of the mentor’s project career path to produce aspirational domain similarity. Then finally equation (5) averages both aspirational skill and domain similarity to produce final user’s aspirational similarity in project career path of the candidate mentor.
[070] Similarly, percentage of role aspiration information of the user matching with project or role career path comprised in the past career path of the one or more candidate mentors is determined in accordance with the following equation provided by way of example below:
(6)
wherein aspsimr (σ(r), p) is a similarity measure of aspiration of the user and the role career path of the at least one candidate mentor, wherein is the role career path of the at least one candidate mentor and p is the aspiration of the user respectively, wherein T is a total duration of wherein σi(p).ro is a role of the
ith project event of the at least one candidate mentor, and wherein p.r is an aspiration role of the user.
[071] More specifically, above equation (6) enables the system 100 to compute the candidate mentor’s total durations of project allocations in user’s aspiration role and is further normalized it by total duration of the mentor’s profile to produce the aspirational role similarity.
[072] Similarly, percentage of skills aspiration information of the user matching training career path comprised in the past career path of the one or more candidate mentors is determined in accordance with the following equation provided by way of example below:

wherein aspsim is a similarity measure of aspiration of the user and the
training career path of the at least one candidate mentor, wherein is the training career path of the at least one candidate mentor and p is the aspiration of the user

respectively, wherein p. CL is a set of aspirational skill clusters, wherein c is an instance of p. CL, and wherein t2sc is a skill cluster associated with training ‘t.
[073] More specifically, above equation (7) enables the system 100 to aggregate the maximum skill cluster similarity between any skill clusters mapping to a training event of mentor to a skill cluster mentioned by user. Then finally the system 100 normalizes the aggregated value by the number of training events in candidate mentor’s project career path to produce aspirational skill cluster similarity of user’s aspirational skill clusters to candidate mentors training career path.
[074] Similarly, percentage of skills aspiration information of the user matching certification career path comprised in the past career path of the one or more candidate mentors is determined and expressed as aspsimc(σ(c),p).
[075] And the second comparison resulting in the second personalized matching career path recommendation data is obtained by taking average or combination of equations (5) to (7) along with the percentage computation of certification and is expressed by way of following exemplary equation:

[076] A list of optimal career paths of the one or more candidate mentors is generated and recommended based on the first personalized matching career path recommendation data (equations (1) to (4)) and the second personalized matching career path recommendation data (equations (1) to (4)), wherein the list of optimal career paths of the one or more candidate mentors is indicative of the at least one optimal career path with the time-bound action plan for recommendation to the user. More specifically, the first personalized matching career path recommendation data and the second personalized matching career path recommendation data are combined to generate the at least one optimal career path and is recommended thereof.
[077] The above steps of domain knowledge driven approach are better understood by way of following description:

[078] As discussed above, providing career path related advice is as important part of mentoring. Human experts (particularly, senior executives and managers) have deep experiential insights into pros and cons of different career choices. Understanding of the psychological strengths and weaknesses of a specific employee allows a mentor to personalize the career recommendations. Such subjective and often not well articulated domain knowledge are captured by the system 100 as part of the similarity measures. Given a CPσ, it is denoted by σ(p), as the projections of σ for each of the 4 event types. For instance, The task of computing similarity sim(σ1, σ2) is broken down and depicted by way of non-construing expressions between two given into computing event type specific similarities, as in expression
(4). Here, is the similarity between the projected subsequences
of σ1 and σ2 restricted to events of type project_alloc. The other (sub) similarity measures are understood analogously. Computations of these (sub) similarity methods are defined as below:

CPs containing only events of type project_alloc. Each pi (and qj) is a tuple of the form: (d,dm,L), where d is the duration of allocation, dm is the industrial domain of the project, and L is the set of skill clusters used in the project. Let T1 = be the sum of the durations of all projects in is defined
similarly. Let C1, C2 denote the set consisting of all skill clusters used in σ(p)1 and respectively. (equation (1)) computes the similarity of two
sequences of project_alloc events as the fraction of the sums of durations of projects using common skill clusters. The method to compute similarity between two role sequences is similar. Let R1, R2 denote the set consisting of all roles occurring in respectively. Equation (2) shows the
similarity computation. Let G1, G2 denote the set consisting of all trainings occurring in respectively. Present disclosure implements Jaccard
similarity between the sequences of trainings (Equation. (3)). A similar computation is for similarity between certification sequences. The similarity





measure aspsim(σ, p) between a CP σ and an aspiration p is now being defined. It is assumed that an n × n similarity matrixM , where
1 denotes the similarity between skill cluster, where SK is set of all skill


clusters. It is further assumes that the function t2sc gives the skill cluster for any particular training; e.g., t2sc(Foundationf Java)= JAVA. As earlier, aspsim is split into sub-similarities over the 4 types of event subsequences (Equation (8)). Let C1 = p. CL and C2 is the union of all skill cluster sets in each project_alloc event in σ(p). Let T be the sum of durations of all project_alloc events in σ(p). Equations (5) to (7) define the required similarity computations. Upon solving equations (1) to (8), a model CP (or also referred as an optimal career path) is identified for the given query employee’s and her aspiration pq. The database
D of CPs (of other employees) is queried and a subset of ideal Ps (e.g., a list of career paths associated with candidate mentors), such that within every model CPσ ideal, there is (i) an initial subsequence a2 which is very similar to σq; and (ii) a later subsequence a3 (which occurs after a2), which is very similar to the aspiration p. Simple sliding window algorithms are designed to find such idealCPs. For (i), pseudo code of the sliding window algorithm is depicted below slides a window W over a given and returns W if the similaritysim(σq, W) of the window
with σq is sufficiently high (i.e., above a user-specified threshold). The algorithm getsuffix for (ii) is similar, except that it slides the window only on the subsequence after a2 and uses the similarity measure aspsim(p,W). The algorithms try windows of different lengths over a range. The above description can be further understood by way of below exemplary pseudo code provided for finding best prefix and suffix CP, wherein prefix CP is matched with user’s CP and prefix CP is a point is before the match, and suffix CP is matched with user’s aspiration and suffix CP is a point after the match.
[080] Pseudo code for finding best prefix and suffix CP (also referred as ‘Optimally matched career path’):



[081] As can be seen through above description and experimented conducted by the embodiments of the present disclosure, domain-knowledge driven similarity approach tends to ignore/disregard a “semantic” similarity between information elements. For example, it assumes zero similarity between DEVELOPER and MODULELEADER roles, between Java and Oracle skill clusters, and between Banking and Insurance domains. However, these pairs are known to be somewhat similar, and this knowledge should be used in computing CP similarity. Another constraint with the domain-knowledge driven similarity is that it ignores the temporal information present in the CP as a sequence of events. For instance, it will treat the CPs σx = (DEVELOPER, TESTER, PROJECTLEADER) and σ2 = (TESTER, DEVELOPER, PROJECTLEADER) as identical, whereas one may reasonably argue that they are somewhat different because of the different order of roles. To overcome these constraints, present disclosure and its systems (e.g., system 100) implements a deep learning technique (e.g., a long short-term memory (LSTM) neural network.
[082] Below is an example of an optimal career path generated and recommendation for the user based on the aspiration information of the user: Table 1: depicting aspirational information of the user.

Aspiration role Technical Architect
Aspirational Domain Banking and Financial Solutions

Aspirational Skill Clusters Java, Oracle data base

Table 2: Optimal career path generated and recommendation for the user

Project Proj1 (java, BFS) Proj2 (java, Insurance) Proj3 (Java,
Allocation oracle, BFS
Role Developer Designer Module Technical Architect
Leader
Training TCS OOPS
Financial Solutions Design Architect
Certification OCJP
Calendar Apr 2016 Oct Apr Oct Apr Oct Apr Oct
Time 2016 2017 2017 2018 2018 2019 2019
Dataset:
[083] Dataset comprises of 24,184 CPs of employees from an organization (e.g., say a large multinational IT company), containing 250,470 events. Below Table 1 shows the distribution of events of each type. The projects are in 19 different industry segments (domains). On average, a CP contained projects 2.87 distinct domains. Another Table shows summaries of the data for top 4 domains, roles, and skill clusters.
Table 3 depicts summary of CP dataset - 1

project _alloc role _set training certifications
#events 108869 38796 100474 2331
average #events/ CP 5.50 1.60 4.15 0.09
total duration 14936539 5578974 100474 2311
average duration/CP 617 230 4.1 0.09

Table 4 depicts summary of CP dataset – 2

Banking Insurance Sales Retail
#projects 9478 3120 1574 1115
Total duration 10572774 2698006 478354 252864
average #events/ CP 20090 5024 1646 1020
Average duration/ CP 526.3 537.0 290.6 247.9
Standard deviation duration/ CP 402.1 427.4 316.4 235.1
DEV PROC_EX DATA EX M LEAD
#projects 5164 57 169 1166
Total duration 1886930 444611 547034 258665
average #events/ CP 4007 903 869 712
Average duration/ CP 470.9 492.4 629.5 363.3
Standard deviation duration/ CP 328.3 311.1 323.0 307.3
Java .NET SQL SVR ORADB
#projects 4279 2955 1585 1600
Total duration 7189380 11373309 106225 3800001
average #events/ CP 12156 11468 10841 8706

Average duration/ CP 591.4 991.74 644.3 436.5
Standard deviation duration/CP 716.6 1634.8 905.4 551.5
Experimental Results:
[084] An experiment was conducted by the present disclosure where aspiration was automatically identified for an employee. 119 “highly experienced” query employees having more than 10 years of experience were randomly selected. Let σq denote the full CP of one such employee. For each such employee, the present disclosure considered the last role, the last domain and the skill clusters used in the last project as the aspiration pq. CP pq was truncated and only the CP corresponding to the initial 30% duration was retained; σq and σq' denote the truncated prefix and suffix CP for employee q, so that σq.σq' = σq. Then the best matching ideal CP a was determined for each employee and the associated aspiration using various methods as described in the present disclosure. The question is: how do we evaluate the quality of the recommended ideal CP σ? [085] Present disclosure used the following measures: [086] Precision k: System 100 computed precision k, which measures for how many of the 119 query CPs, the original CP appeared within the top k most similar CPs identified using each similarity method. Below table depicts results for Precision k.
Table 5: Precision k results

k B M1 M2
1 0.89 0.88 0.39
2 0.95 0.94 0.54
3 0.99 0.96 0.68
5 1.00 0.99 0.74
10 1.00 1.0 0.80

[087] Self-rank: For the CP of a given query employee q, all CPs in the historical database were ranked (using a specific method), and the rank of the full CP sigmaq was computed. In the experiments, only an initial 30% part the entire CP of q was used to find the ideal CP for it, meaning the full CP may or may not be the ideal CP. This measure is referred as self-rank of q. If a method produces a low self-rank for q (averaged over several q) then it implies that the method is better.
[088] Achievement duration: By construction, the full has
achieved the aspiration pq. If the ideal CP “achieves” the aspiration in less time than the time taken by sigmaq to achieve pq, then the ideal CP is indeed better than the path actually followed by q, and hence the recommendation is successful; otherwise, the recommendation has not been that successful. The number of successful recommendations for the 100 chosen CPs is second evaluation measure by the present disclosure.
[089] The achievement duration of an aspiration in a given
of employee e defined as follows. Let the set CL consist of p skill clusters. Let tx denote the earliest time at which the employee e gets assigned to a project in domain d, and where her role is r and where some of the skill clusters from CL is being used, and where all the other skill clusters in CL were used in some previous project in which e worked. Then the achievement duration Ae for employee e is the duration between the first (initial) time instant of and t1 FIG. 4 shows an example CP, and marks the time instant where the aspiration p = has been achieved, thereby making the achievement duration to be 547 days with respect to the beginning of the CP. More specifically, FIG. 4, with reference to FIGS. 1 through 3, depicts an example for achievement duration in accordance with an embodiment of the present disclosure. It is assumed that the system 100 not only returns the ideal CP a, but also its prefix (and the remaining suffix which best matched the query Thus, the
achievement durations Aq and Aideal need to be computed for the truncated suffix

as well as for the suffix σ1 of the ideal CP. The difference is the
measure of success; positive and large values indicate that the ideal CP has achieved the aspiration much more quickly than the actual CP followed by the query employee.
[090] Average similarity: This is the average of the similarity between σ0 and σ'q and similarity between This measure checks how similar the
prefixes (and suffixes) of the ideal CP and query CP are. Table shows the comparison of methods of the present disclosure using P@k; B, M1, M2 refer to the baseline method, domain-knowledge driven method and embedding-based edit distance method respectively. Method M1 clearly outperforms both B and M2. Table shows the comparison of methods of the present disclosure using self-rank and average similarity. Method M1 clearly outperforms both B and M2. The number of successful recommendations (where ideal CP had a shorted achievement duration) for methods B, M1, M2 are 16 (13.4%), 17 (14%) and 6 (5.0%), showing the superiority of M1.
[091] Below table depicts results for Self-rank and Average similarity: Table 6: Self-rank and Average similarity

Self-rank Average similarity

B M1 M2 B M1 M2
average 1.24 1.25 7.63 0.37 0.34 0.26
median 1.0 1.0 2.0 0.37 0.35 0.26
STDEV 0.93 0.95 12.6 0.10 0.13 0.08
[092] The written description describes the subject matter herein to enable any person skilled in the art to make and use the embodiments. The scope of the subject matter embodiments is defined by the claims and may include other modifications that occur to those skilled in the art. Such other modifications are intended to be within the scope of the claims if they have similar elements that do not differ from the literal language of the claims or if they include equivalent elements with insubstantial differences from the literal language of the claims.

[093] It is to be understood that the scope of the protection is extended to such a program and in addition to a computer-readable means having a message therein; such computer-readable storage means contain program-code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The hardware device can be any kind of device which can be programmed including e.g. any kind of computer like a server or a personal computer, or the like, or any combination thereof. The device may also include means which could be e.g. hardware means like e.g. an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination of hardware and software means, e.g. an ASIC and an FPGA, or at least one microprocessor and at least one memory with software processing components located therein. Thus, the means can include both hardware means and software means. The method embodiments described herein could be implemented in hardware and software. The device may also include software means. Alternatively, the embodiments may be implemented on different hardware devices, e.g. using a plurality of CPUs.
[094] The embodiments herein can comprise hardware and software elements. The embodiments that are implemented in software include but are not limited to, firmware, resident software, microcode, etc. The functions performed by various components described herein may be implemented in other components or combinations of other components. For the purposes of this description, a computer-usable or computer readable medium can be any apparatus that can comprise, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
[095] The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed. These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are

appropriately performed. Alternatives (including equivalents, extensions,
variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope of the disclosed embodiments. Also, the words “comprising,” “having,” “containing,” and “including,” and other similar forms are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise.
[096] Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term “computer-readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., be non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.
[097] It is intended that the disclosure and examples be considered as exemplary only, with a true scope of disclosed embodiments being indicated by the following claims.

We Claim:
1. A processor implemented method, comprising:
obtaining (202), via one or more hardware processors, information pertaining to a career path and an aspiration information specific to a user associated with an entity, wherein the career path comprises one or more past events associated with the user;
generating and recommending (204), via the one or more hardware processors, at least one optimal career path with a time-bound action plan using the obtained information, wherein the step of generating and recommending the at least one optimal career path with the time-bound action plan comprises:
obtaining (204a), via a deep learning neural network executed by the one or more hardware processors, one or more career affecting events pertaining to the user and one or more candidate mentors associated with the entity;
learning (204b), via the deep learning neural network executed by the one or more hardware processors, an embedding of each of the one or more career affecting events for a plurality of users associated with the entity, wherein the learnt embedding is indicative of an embedded mathematical representation of relationship among each of the one or more career affecting events;
performing in the embedded mathematical representation (204c), using a career edit distance technique executed by the one or more hardware processors, a first comparison of (i) the career path of the user with (ii) a past career path of the one or more candidate mentors to obtain a first personalized matching career path recommendation data, wherein the first personalized matching career path recommendation data comprises an optimally matched career path of the one or more candidate mentors based on the career path of the user;
performing in the embedded mathematical representation (204d), using a cosine similarity technique executed by the one or more hardware

processors, a second comparison of (i) the aspirational information of the user with (ii) the past career path of the one or more candidate mentors to obtain a second personalized matching career path recommendation data; and
generating (204e), via the one or more hardware processors, a list of optimal career paths of the one or more candidate mentors based on the first personalized matching career path recommendation data and the second personalized matching career path recommendation data, wherein the list of optimal career paths of the one or more candidate mentors is indicative of the at least one optimal career path with the time-bound action plan for recommendation to the user.
2. The processor implemented method as claimed in claim 1, wherein the at least one optimal career path comprises at least one of (i) a sequence of one or more roles, (ii) a sequence of one or more projects, (iii) at least one training course, and (iv) at least one certification, and wherein each of the one or more role and the one or more projects comprise a recommended career timeline.
3. The processor implemented method of claim 1, wherein the one or more career affecting events comprise one or more projects being allocated to the user, one or more roles being assigned to the user, one or more training courses attended or to be attended by the user, and associated one or more certifications thereof.
4. The processor implemented method as claimed in claim 1, wherein the step of generating and recommending, via the one or more hardware processors, at least one optimal career path, is based on a domain knowledge driven technique comprising:
obtaining, via the one or more hardware processors, the one or more career affecting events pertaining to the user and the one or more candidate mentors associated with the entity;

performing based on the one or more career affecting events, via the one or more hardware processors, a first comparison of (i) a career path of the user with (ii) a past career path of the one or more candidate mentors to obtain a first personalized matching career path recommendation data,
wherein the first personalized matching career path recommendation data comprises information on (i) percentage of a project career path of the user matching a past project career path comprised in the past career path of the one or more candidate mentors, (ii) percentage of a role career path of the user matching a past role career path comprised in the past career path of the one or more candidate mentors; (iii) percentage of training attended or training in progress, and certifications received by the user matching training courses and certifications comprised in the past career path of the one or more candidate mentors,
wherein the percentage of a project career path of the user matching a past project career path comprised in the past career path of the one or more candidate mentors is computed in accordance with an equation:

wherein is a similarity measure between at least two
project career paths, wherein is a projection of a project career path of the user, wherein is a projection of the project career path of at least one candidate
mentor, wherein 7\ is a total duration of wherein T2 is a total duration of
wherein DOM is an intersection of one or more domains in and wherein
m is number of events in i is an iterator,is a duration in an ith project
event of , wherein n is number of events in wherein qJm d is a duration in
an jth project event of wherein C is an intersection of skills inand

wherein C is an instance of C, wherein pi. Lis a set of skill clusters associated with the ith project event of the user, wherein qj. L is a set of skill clusters associated with the jth project event of the at least one candidate mentor,
wherein the percentage of a role career path of the user matching a past role career path comprised in the past career path of the one or more candidate mentors is computed in accordance with an equation:

wherein is a similarity measure between at least two role career
paths, wherein is a projection of a role career path of the user, wherein is a projection of the role career path of at least one candidate mentor, wherein T1 is a total duration of σ1(r), wherein T2 is a total duration of σ(r)2, wherein R is an
intersection of roles in σ1(r) and σ2(r), wherein r is an instance of R, wherein pi.ro is a role mention in the ith project event, qj. r ois a role mention in the jth project event,
wherein the percentage of training attended or being attending by the user matching training courses comprised in the past career path of the one or more candidate mentors is computed in accordance with an equation:
is a similarity measure between at least two
training career paths, wherein is a projection of a training career path of the user, wherein is a projection of a training career path of at least one candidate mentor, wherein C1 is a set of trainings in wherein C2 is a set of trainings in

wherein percentage of certifications received by the user matching certifications comprised in the past career path of the one or more candidate mentors
is expressed as wherein is a similarity measure
between at least two certification career paths, wherein is a projection of a

certification career path of the user, wherein σ2(c) is a projection of a certification career path of at least one candidate mentor, and
wherein the first personalized matching career path recommendation data is computed in accordance with an equation:

performing, via the one or more hardware processors, a second comparison of (i) the aspirational information of the user with (ii) the past career path of the one or more candidate mentors to obtain a second personalized matching career path recommendation data, wherein the second personalized matching career path recommendation data comprises information on (a) percentage of the skill and domain aspiration information of the user matching project career path comprised in the past career path of the one or more candidate mentors, (b) percentage of role aspiration information of the user matching with project career path or role career path comprised in the past career path of the one or more candidate mentors, (c) percentage of skills aspiration information of the user matching a training career path comprised in the past career path of the one or more candidate mentors, and (d) percentage of skills aspiration information of the user matching a certification career path comprised in the past career path of the one or more candidate mentors,
wherein the percentage of the skill and domain aspiration information of the user matching project career path comprised in the past career path of the one or more candidate mentors is computed in accordance with an equation:

wherein aspsimp (σ(p), p) is a similarity measure of
aspiration of the user and the project career path of the at least one candidate mentor, wherein is the project career path of the at least one candidate mentor and p is the aspiration of the user, wherein T is a total duration of , wherein C1 is a set of skill clusters in and C2 is a set of skill clusters in p respectively, wherein x

and y are instances of C1 and C2 respectively, wherein Mx,y is a similarity measure
between x and y, wherein is a set of skill clusters in the ith project event of
the at least one candidate mentor, wherein is a duration of the ith project
event of whereinis a domain of the ith project event of the at least
one candidate mentor, wherein p. d is an aspiration domain of the user,
wherein the percentage of role aspiration information of the user matching with project career path or role career path comprised in the past career path of the one or more candidate mentors is computed in accordance with an equation:

wherein is a similarity measure of aspiration of the user
and the role career path of the at least one candidate mentor, wherein is the role career path of the at least one candidate mentor and p is the aspiration of the user
respectively, wherein T is a total duration of , wherein ro is a role of the
ith project event of the at least one candidate mentor, wherein p.r is an aspiration role of the user,
wherein the percentage of skills aspiration information of the user matching a training career path comprised in the past career path of the one or more candidate mentors is computed in accordance with an equation:

wherein is a similarity measure of aspiration of the user
and the training career path of the at least one candidate mentor, wherein σ(t) is the training career path of the at least one candidate mentor and p is the aspiration of the user respectively, wherein p.CLis a set of aspirational skill clusters, wherein c is an instance of p.CL, wherein t2sc is a skill cluster associated with training 't,
wherein, the percentage of skills aspiration information of the user matching a certification career path comprised in the past career path of the one or more candidate mentors is expressed as and

wherein the second personalized matching career path recommendation data is computed in accordance with an equation:

generating a list of optimal career paths of the one or more candidate mentors based on the first personalized matching career path recommendation data and the second personalized matching career path recommendation data, wherein the list of optimal career paths of the one or more candidate mentors is indicative of the at least one optimal career path with the time-bound action plan for recommendation to the user.
5. A system (100), comprising:
a memory (102) storing instructions;
one or more communication interfaces (106); and
one or more hardware processors (104) coupled to the memory (102) via the one or more communication interfaces (106), wherein the one or more hardware processors (104) are configured by the instructions to:
obtain information pertaining to a career path and an aspiration information specific to a user associated with an entity, wherein the career path comprises one or more past events associated with the user;
generate and recommend at least one optimal career path with a time-bound action plan using the obtained information, wherein the at least one optimal career path with the time-bound action plan is generated and recommended by:
obtaining, via a deep learning neural network executed by the one or
more hardware processors, one or more career affecting events pertaining
to the user and one or more candidate mentors associated with the entity;
learning, via the deep learning neural network executed by the one
or more hardware processors, an embedding of each of the one or more
career affecting events for a plurality of users associated with the entity,
wherein the learnt embedding is indicative of an embedded mathematical

representation of relationship among each of the one or more career affecting events;
performing in the embedded mathematical representation, using a career edit distance technique executed by the one or more hardware processors, a first comparison of (i) the career path of the user with (ii) a past career path of the one or more candidate mentors to obtain a first personalized matching career path recommendation data, wherein the first personalized matching career path recommendation data comprises an optimally matched career path of the one or more candidate mentors based on the career path of the user;
performing in the embedded mathematical representation, using a cosine similarity technique executed by the one or more hardware processors, a second comparison of (i) the aspirational information of the user with (ii) the past career path of the one or more candidate mentors to obtain a second personalized matching career path recommendation data; and
generating, via the one or more hardware processors, a list of optimal career paths of the one or more candidate mentors based on the first personalized matching career path recommendation data and the second personalized matching career path recommendation data, wherein the list of optimal career paths of the one or more candidate mentors is indicative of the at least one optimal career path with the time-bound action plan for recommendation to the user.
6. The system as claimed in claim 5, wherein the at least one optimal career
path comprises at least one of (i) a sequence of one or more roles, (ii) a sequence of one or more projects, (iii) at least one training course, and (iv) at least one certification, and wherein each of the one or more role and the one or more projects comprise a recommended career timeline.

7. The system as claimed in claim 5, wherein the one or more career affecting events comprise one or more projects being allocated to the user, one or more roles being assigned to the user, one or more training courses attended or to be attended by the user, and associated one or more certifications thereof.
8. The system as claimed in claim 5, wherein the at least one optimal career path is generated and recommended using a domain knowledge driven technique by:
obtaining, via the one or more hardware processors, the one or more career affecting events pertaining to the user and the one or more candidate mentors associated with the entity;
performing based on the one or more career affecting events, via the one or more hardware processors, a first comparison of (i) a career path of the user with (ii) a past career path of the one or more candidate mentors to obtain a first personalized matching career path recommendation data,
wherein the first personalized matching career path recommendation data comprises information on (i) percentage of a project career path of the user matching a past project career path comprised in the past career path of the one or more candidate mentors, (ii) percentage of a role career path of the user matching a past role career path comprised in the past career path of the one or more candidate mentors; (iii) percentage of training attended or training in progress, and certifications received by the user matching training courses and certifications comprised in the past career path of the one or more candidate mentors,
wherein the percentage of a project career path of the user matching a past project career path comprised in the past career path of the one or more candidate mentors is computed in accordance with an equation:


wherein is a similarity measure between at least two
project career paths, wherein is a projection of a project career path of the user, wherein is a projection of the project career path of at least one candidate
mentor, wherein T1 is a total duration of wherein T2 is a total duration of wherein DOM is an intersection of one or more domains in and σ2(p), wherein m is number of events in i is an iterator, pi.d is a duration in an ith project
event of wherein n is number of events in wherein qj.d is a duration in
an jth project event of wherein is an intersection of skills in and
wherein C is an instance of C, wherein pi.L is a set of skill clusters associated with the ith project event of the user, wherein qj.L is a set of skill clusters associated with the jth project event of the at least one candidate mentor,
wherein the percentage of a role career path of the user matching a past role career path comprised in the past career path of the one or more candidate mentors is computed in accordance with an equation:

wherein simr (σ1(r), σ2(r) )is a similarity measure between at least two role
career paths, wherein is a projection of a role career path of the user, wherein
is a projection of the role career path of at least one candidate mentor, wherein
T1 is a total duration of wherein T2 is a total duration of wherein R is an
intersection of roles in and wherein r is an instance of �, wherein pi.ro

is a role mention in the ith project event, qj.ro is a role mention in the jth project event,
wherein the percentage of training attended or being attending by the user matching training courses comprised in the past career path of the one or more candidate mentors is computed in accordance with an equation:

wherein simt (σ(t)1, σ(t)2) is a similarity measure between at least two
training career paths, wherein is a projection of a training career path of the user, wherein is a projection of a training career path of at least one candidate mentor, wherein C1 is a set of trainings in , wherein C2 is a set of trainings in

wherein percentage of certifications received by the user matching certifications comprised in the past career path of the one or more candidate mentors
is expressed as wherein is a similarity measure
between at least two certification career paths, wherein is a projection of a
certification career path of the user, wherein is a projection of a certification career path of at least one candidate mentor, and
wherein the first personalized matching career path recommendation data is computed in accordance with an equation:

performing, via the one or more hardware processors, a second comparison of (i) the aspirational information of the user with (ii) the past career path of the one or more candidate mentors to obtain a second personalized matching career path recommendation data, wherein the second personalized matching career path recommendation data comprises information on (a) percentage of the skill and domain aspiration information of the user matching project career path comprised in the past career path of the one or more candidate mentors, (b) percentage of role

aspiration information of the user matching with project career path or role career path comprised in the past career path of the one or more candidate mentors, (c) percentage of skills aspiration information of the user matching a training career path comprised in the past career path of the one or more candidate mentors, and (d) percentage of skills aspiration information of the user matching a certification career path comprised in the past career path of the one or more candidate mentors, wherein the percentage of the skill and domain aspiration information of the user matching project career path comprised in the past career path of the one or more candidate mentors is computed in accordance with an equation:

wherein is a similarity measure of
aspiration of the user and the project career path of the at least one candidate mentor, wherein is the project career path of the at least one candidate mentor and p is the aspiration of the user, wherein T is a total duration of , wherein c1 is a set of skill clusters in and C2 is a set of skill clusters in p respectively, wherein x and y are instances of C1 and C2 respectively, wherein is a similarity measure
between x and y, wherein is a set of skill clusters in the ith project event of
the at least one candidate mentor, wherein is a duration of the ith project
event of , wherein dm is a domain of the ith project event of the at least
one candidate mentor, wherein p. d is an aspiration domain of the user,
wherein the percentage of role aspiration information of the user matching with project career path or role career path comprised in the past career path of the one or more candidate mentors is computed in accordance with an equation:

wherein is a similarity measure of aspiration of the user
and the role career path of the at least one candidate mentor, wherein is the role career path of the at least one candidate mentor and p is the aspiration of the user

respectively, wherein T is a total duration of wherein is a role of the
ith project event of the at least one candidate mentor, wherein p.r is an aspiration role of the user,
wherein the percentage of skills aspiration information of the user matching a training career path comprised in the past career path of the one or more candidate mentors is computed in accordance with an equation:

wherein aspsimt(σ(t), p) is a similarity measure of aspiration of the user and the training career path of the at least one candidate mentor, wherein is the training career path of the at least one candidate mentor and p is the aspiration of the user respectively, wherein p. CL is a set of aspirational skill clusters, wherein σ(t) is an instance of p.CL, wherein t2sc is a skill cluster associated with training ‘t,
wherein, the percentage of skills aspiration information of the user matching a certification career path comprised in the past career path of the one or more candidate mentors is expressed as and
wherein the second personalized matching career path recommendation data is computed in accordance with an equation:

generating a list of optimal career paths of the one or more candidate mentors based on the first personalized matching career path recommendation data and the second personalized matching career path recommendation data, wherein the list of optimal career paths of the one or more candidate mentors is indicative of the at least one optimal career path with the time-bound action plan for recommendation to the user.