FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003
COMPLETE SPECIFICATION (See Section 10 and Rule 13)
Title of invention:
METHOD AND SYSTEM FOR MACHINE LEARNING BASED ROLE
FITMENT ASSESSMENT
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 the field of machine learning and, more particular, to a method and system for machine learning based role fitment assessment.
BACKGROUND
[002] Behavioral profiling is an important aspect to assess and improve a person’s behavioral tendencies, preferred work styles, communication abilities and decision-making styles. Behavioral profiling is also useful for organizations to assess role fitment of a candidate corresponding to a role, wherein the role fitment is a combination of cognitive ability and behavioral style.
[003] Conventional methods mainly depend on expert opinion of the candidates in order to arrive at a role fitment decision. However, it is challenging to scale and prone due to human error and bias. Hence there is a challenge in developing a scalable and sustainable role fitment model to identify the potential pool proactively, ensuring a robust talent pipeline and planned organizational growth and continuity.
SUMMARY [004] 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 embodiment, a method for machine learning based role fitment assessment is provided. The method includes receiving, by one or more hardware processors, an assessment data pertaining to a user, wherein the assessment data comprises a textual data, a survey data and an audiovisual data, wherein the user is associated with a competency profile. Further, the method includes computing by the one or more hardware processors, a plurality of textual features based on the textual data by a linguistic text analysis technique, wherein the plurality of textual features comprises a stand-out factor, a drive for result value, a reasoning ability value and a communication skill value. Furthermore, the method includes computing by the

one or more hardware processors, a unified text based competency value based on the plurality of textual features and a corresponding predetermined weight associated with each of the plurality of textual features by a linear regression model. Furthermore, the method includes simultaneously computing by the one or more hardware processors, a plurality of audiovisual features from the audiovisual data by a machine learning technique, wherein the plurality of audiovisual features comprises a confidence value, an emotion value, a modulation score and a speaking quality score. Furthermore, the method includes simultaneously computing by the one or more hardware processors, a plurality of dimensions of competency based on the survey data, wherein the survey data is obtained based on the competency profile of the user. Finally, the method includes computing by the one or more hardware processors, a role fitment value of the user based on the unified text based competency score, the plurality of audiovisual features and the plurality of survey features by a random forest classification.
[005] In another aspect, a system for machine learning based role fitment assessment is provided. The system includes at least one memory storing programmed instructions, one or more Input /Output (I/O) interfaces, and one or more hardware processors operatively coupled to the at least one memory, wherein the one or more hardware processors are configured by the programmed instructions to receive an assessment data pertaining to a user, wherein the assessment data comprises a textual data, a survey data and an audiovisual data, wherein the user is associated with a competency profile. Further, the one or more hardware processors are configured by the programmed instructions to compute a plurality of textual features based on the textual data by a linguistic text analysis technique, wherein the plurality of textual features comprises a stand-out factor, a drive for result value, a reasoning ability value and a communication skill value. Furthermore, the one or more hardware processors are configured by the programmed instructions to compute a unified text based competency value based on the plurality of textual features and a corresponding predetermined weight associated with each of the plurality of textual features by a linear regression model. Furthermore, the one or more hardware processors are configured by the

programmed instructions to simultaneously compute a plurality of audiovisual features from the audiovisual data by a machine learning technique, wherein the plurality of audiovisual features comprises a confidence value, an emotion value, a modulation score and a speaking quality score. Furthermore, the one or more hardware processors are configured by the programmed instructions to simultaneously compute a plurality of dimensions of competency based on the survey data, wherein the survey data is obtained based on the competency profile of the user. Finally, the one or more hardware processors are configured by the programmed instructions to compute a role fitment value of the user based on the unified text based competency score, the plurality of audiovisual features and the plurality of survey features by a random forest classification.
[006] In yet another aspect, a computer program product including a non-transitory computer-readable medium having embodied therein a computer program for machine learning based role fitment assessment is provided. The computer readable program, when executed on a computing device, causes the computing device to receive an assessment data pertaining to a user, wherein the assessment data comprises a textual data, a survey data and an audiovisual data, wherein the user is associated with a competency profile. Further, the computer readable program, when executed on a computing device, causes the computing device to compute a plurality of textual features based on the textual data by a linguistic text analysis technique, wherein the plurality of textual features comprises a stand-out factor, a drive for result value, a reasoning ability value and a communication skill value. Furthermore, the computer readable program, when executed on a computing device, causes the computing device to compute a unified text based competency value based on the plurality of textual features and a corresponding predetermined weight associated with each of the plurality of textual features by a linear regression model. Furthermore, the computer readable program, when executed on a computing device, causes the computing device to simultaneously compute a plurality of audiovisual features from the audiovisual data by a machine learning technique, wherein the plurality of audiovisual features comprises a confidence value, an emotion value, a modulation score and a speaking

quality score. Furthermore, the computer readable program, when executed on a computing device, causes the computing device to simultaneously compute a plurality of dimensions of competency based on the survey data, wherein the survey data is obtained based on the competency profile of the user. Finally, the computer readable program, when executed on a computing device, causes the computing device to compute a role fitment value of the user based on the unified text based competency score, the plurality of audiovisual features and the plurality of survey features by a random forest classification.
[007] 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
[008] 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:
[009] FIG. 1 is a functional block diagram of a system for machine learning based role fitment assessment, in accordance with some embodiments of the present disclosure.
[0010] FIGS. 2 is an exemplary flow diagram illustrating a method for machine learning based role fitment assessment, implemented by the system of FIG. 1, in accordance with some embodiments of the present disclosure.
[0011] FIG. 3 is a functional block diagram for unified text based competency value computation for the processor implemented method for machine learning based role fitment assessment implemented by the system of FIG. 1, in accordance with some embodiments of the present disclosure.
[0012] FIG. 4 is a functional block diagram for audio-visual feature computation for the processor implemented method for machine learning based role fitment assessment implemented by the system of FIG. 1, in accordance with some embodiments of the present disclosure.

[0013] FIG. 5 is overview of an example architecture for the processor implemented method for machine learning based role fitment assessment implemented by the system of FIG. 1, in accordance with some embodiments of the present disclosure.
DETAILED DESCRIPTION OF EMBODIMENTS [0014] 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 spirit and scope of the disclosed embodiments. [0015] Embodiments herein provide a method and system for machine learning based role fitment assessment for assessing competency of a user corresponding to a role. The present system and method utilize advances in artificial intelligence based speech, vision and language processing technologies to discover hidden potential of users from multi-modal data. Initially, the system receives an assessment data pertaining to the user under assessment. The assessment data includes a textual data, a survey data and an audiovisual data, wherein the user is associated with a competency profile. Further, a plurality of textual features are computed based on the textual data by a linguistic text analysis, wherein the plurality of textual features includes a stand-out factor, a drive for result value, a reasoning ability value and a communication skill value. After computing the plurality of textual features, a unified text based competency value is computed based on the plurality of textual features and a corresponding predetermined weight associated with each of the plurality of textual features by a linear regression model. A plurality of audiovisual features is simultaneously computed from the audiovisual data by a machine learning technique. A plurality of dimensions of competency is simultaneously computed based on the survey data, wherein the survey data is obtained based on the competency profile of the user. Finally, a role fitment value

of the user is computed based on the unified text based competency score, the plurality of audiovisual features and the plurality of survey features by a random forest classification.
[0016] Referring now to the drawings, and more particularly to FIGS. 1 through 5, 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.
[0017] FIG. 1 is a functional block diagram of a system 100 for machine learning based role fitment assessment, according to some embodiments of the present disclosure. The system 100 includes or is otherwise in communication with hardware processors 102, at least one memory such as a memory 104, an I/O interface 112. The hardware processors 102, memory 104, and the Input /Output (I/O) interface 112 may be coupled by a system bus such as a system bus 108 or a similar mechanism. In an embodiment, the hardware processors 102 can be one or more hardware processors.
[0018] The I/O interface 112 may include a variety of software and hardware interfaces, for example, a web interface, a graphical user interface, and the like. The I/O interface 112 may include a variety of software and hardware interfaces, for example, interfaces for peripheral device(s), such as a keyboard, a mouse, an external memory, a printer and the like. Further, the I/O interface 112 may enable the system 100 to communicate with other devices, such as web servers, and external databases.
[0019] The I/O interface 112 can facilitate multiple communications within a wide variety of networks and protocol types, including wired networks, for example, local area network (LAN), cable, etc., and wireless networks, such as Wireless LAN (WLAN), cellular, or satellite. For the purpose, the I/O interface 112 may include one or more ports for connecting several computing systems with one another or to another server computer. The I/O interface 112 may include one or more ports for connecting several devices to one another or to another server.

[0020] The one or more hardware processors 102 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, node machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the one or more hardware processors 102 is configured to fetch and execute computer-readable instructions stored in the memory 104.
[0021] The memory 104 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, the memory 104 includes a plurality of modules 106. The memory 104 also includes a data repository (or repository) 110 for storing data processed, received, and generated by the plurality of modules 106. The repository 110 further includes a knowledge repository, error repository and a metadata repository.
[0022] The plurality of modules 106 include programs or coded instructions that supplement applications or functions performed by the system 100 for machine learning based role fitment assessment. The plurality of modules 106, amongst other things, can include routines, programs, objects, components, and data structures, which performs particular tasks or implement particular abstract data types. The plurality of modules 106 may also be used as, signal processor(s), node machine(s), logic circuitries, and/or any other device or component that manipulates signals based on operational instructions. Further, the plurality of modules 106 can be used by hardware, by computer-readable instructions executed by the one or more hardware processors 102, or by a combination thereof. The plurality of modules 106 can include various sub-modules (not shown). The plurality of modules 106 may include computer-readable instructions that supplement applications or functions performed by the system 100 for machine learning based role fitment assessment. In an embodiment, plurality of modules 106 includes a standard scoring and reverse coding module (not shown in FIG. 1), a principal component analysis module (not shown in FIG. 1), a linguistic processing

module (not shown in FIG. 1), a regression module (not shown in FIG. 1), a machine learning module (not shown in FIG. 1) and a random forest classifier module (not shown in FIG. 1).
[0023] The data repository (or repository) 110 may include a plurality of abstracted piece of code for refinement and data that is processed, received, or generated as a result of the execution of the plurality of modules in the module(s) 106.
[0024] Although the data repository 110 is shown internal to the system 100, it will be noted that, in alternate embodiments, the data repository 110 can also be implemented external to the system 100, where the data repository 110 may be stored within a database (not shown in FIG. 1) communicatively coupled to the system 100. The data contained within such external database may be periodically updated. For example, new data may be added into the database (not shown in FIG. 1) and/or existing data may be modified and/or non-useful data may be deleted from the database (not shown in FIG. 1). In one example, the data may be stored in an external system, such as a Lightweight Directory Access Protocol (LDAP) directory and a Relational Database Management System (RDBMS).
[0025] FIGS. 2 is an exemplary flow diagram illustrating a method 200 for machine learning based role fitment assessment implemented by the system of FIG. 1 according to some embodiments of the present disclosure. In an embodiment, the system 100 includes one or more data storage devices or the memory 104 operatively coupled to the one or more hardware processor(s) 102 and is configured to store instructions for execution of steps of the method 200 by the one or more hardware processors 102. The steps of the method 200 of the present disclosure will now be explained with reference to the components or blocks of the system 100 as depicted in FIG. 1 and the steps of flow diagram as depicted in FIG. 2. The method 200 may be described in the general context of computer executable instructions. Generally, computer executable instructions can include routines, programs, objects, components, data structures, procedures, modules, functions, etc., that perform particular functions or implement particular abstract data types. The method 200 may also be practiced in a distributed computing environment where functions are

performed by remote processing devices that are linked through a communication network. The order in which the method 200 is described is not intended to be construed as a limitation, and any number of the described method blocks can be combined in any order to implement the method 200, or an alternative method. Furthermore, the method 200 can be implemented in any suitable hardware, software, firmware, or combination thereof.
[0026] At step 202 of the method 200, the one or more hardware processors 102 are configured by the programmed instructions to receive an assessment data pertaining to a user. The assessment data includes a textual data, a survey data and an audiovisual data. The user is associated with a competency profile. For example, the textual data includes any article or essay written by the user. The survey data includes captured responses of participants (users) on various statements that measure behavioral competencies such as personality, achievement orientation, customer responsiveness etc. The responses are obtained on a 5-point Likert scale (Strongly Agree; Agree; Neutral; Disagree; Strongly Disagree). The audiovisual data includes the recording of role-play performed by the user or the professional experience shared by the user.
[0027] At step 204 of the method 200, the one or more hardware processors 102 are configured by the programmed instructions to compute a plurality of textual features based on the textual data by a linguistic text analysis. The plurality of textual features includes a stand-out factor, a drive for result, a reasoning ability and a communication skill. The drive for result is computed based on a plurality of stylometric feature values. The stand-out factor is computed based on an entropy and a concept uniqueness associated with the textual data. The reasoning ability is computed based on causal relation between a plurality of entities and a plurality of events associated with the textual data, and wherein the communication skill is computed based on a plurality of linguistic features, a plurality of psychological features and a word embedding associated with the textual data.
[0028] In an embodiment, the stand-out factor is computed as explained below: Initially, the textual data pertaining to the user is received. Further, the entropy is computed based on the textual data by (i) generating a plurality of unique

bag-of-words based on the textual data by counting a number of time each of the plurality of unique bag-of-word occurs in the textual data. Each of the plurality of unique bag-of-words is associated with a corresponding weight (ii) obtaining a plurality of smoothed bag-of-words based on the plurality of unique bag-of-words using a linear interpolation smoothing, wherein the linear interpolation smoothing removes zero probability problem when any one of the corresponding weights is zero and (iii) computing the entropy based on each of the plurality of smoothed bag-of-words and the corresponding weight. Further, a plurality of key-phrases is simultaneously computed from the textual data with a corresponding confidence score by a Rapid Automatic Keyword Extraction (RAKE) algorithm. Further a plurality of confident key-phrases is obtained from the plurality of key-phrases based on a predetermined confidence threshold by RAKE. RAKE is a standard phrase extractor tool that returns key phrases along with a score associated with each key phrase. The plurality of key-phrases with the confidence score greater than the predetermined confidence threshold is selected. Furthermore, a corresponding frequency of occurrence of each of the plurality of confident key-phrases are computed by counting a number of times each of the plurality of confident key-phrase occurs in the textual data. Furthermore, a plurality of unique unigrams are obtained by sorting the plurality of confident key-phrases in descending order based on the corresponding frequency of occurrence. Furthermore, a plurality of noun unigrams are obtained based on the plurality of unique unigrams from a WordNet database. Furthermore, a plurality of unigram based multiword phrases are obtained from the plurality of confident key-phrases based on the plurality of noun unigrams. Furthermore, a candidate elaboration list based on the plurality of unigram is obtained based multiword phrases by selecting the plurality of confident key-phrases containing at least one noun unigram from the plurality of noun unigrams. Post generating the candidate elaboration list a plurality of root words are obtained based on the candidate elaboration list by generating a root form for each word in the candidate elaboration list using Porter’s stemming algorithm. Further, a plurality of concepts are obtained by arranging each of the plurality of root words in alphabetical order. Post obtaining the plurality of concepts, a concept uniqueness

value for each of the plurality of concepts are obtained by computing an inverted ratio between a number of documents containing the concept and a total number of documents. For example, key phrases like young billionaire, young billionaires, affluent young billionaire, young billionaire rate, young billionaire cross selling prospect, potential young billionaire, young billionaire target group, young billionaire innovation-all are grouped together under one concept called “young billionaire”. Finally, the stand-out-factor is computed based on the entropy, the concept uniqueness value and a number of concepts. The number of concepts is total number of the plurality of concepts.
[0029] In an embodiment, the entropy of a text refers to the amount of information present in the text. In order to compute Entropy, we have considered that each document d is represented by a bag-of-word as, < (q1, w1), (q2,w2), (q3, w3), …, (qn , wn) > where qi is the ith unique term in document d and wi the corresponding weight computed with respect to a collection of documents C. The entropy of a text T, with N words and n unique ones is defined as given in equation
q1(w1,w2,w3….wn)= 1/N Σni=1 P(wi)* log (P(wi)) ………… (1)
where, P(wi) (i = 1... n) is the probabilistic measure of the ith word in the text T. In order to avoid the problem of zero probabilities, linear interpolation smoothing has been utilized, wherein document weights are smoothed against the set of the documents in the corpus.
[0030] In an embodiment, the concept uniqueness score for a given concept
qacross a document collection C is given in equation (2). CU(q) = log ((N - dfq )/dfq )……………………………..(2) where, N is the total number of documents in the collection. The stand-out factor is obtained by combining the entropy and the conceptual uniqueness scores. The formula for computing the stand-out factor is given in equation (3). stand -out factor (C,d) = Ed * ∑ki=1 CU(qi)………………(3) where Ed is the entropy score of the document d, CU(qi) is the uniqueness score of the concept qi and k is the number of key concepts in a document d.

[0031] In an embodiment, the drive for result is computed as follows. Initially, the textual data pertaining to the user is received. Further, a plurality of stylometric feature values are computed from the textual data using a Linguistic Information and Word Count (LIWC) tool. The plurality of stylometric feature values includes an achiever score, a powerfulness score, a driving factor score, an affect score and an insight score. Finally, the drive for result attribute is computed based on the plurality of stylometric features by computing a weighted average of each of the plurality of stylometric feature scores. For example, if the achiever score, powerfulness score, driving factor score, affect score and an insight score of a text document is [0.34, 0.06, 0.71,0.54, 0.22] then the drive for result score will be = 0.31.
[0032] In an embodiment, the method of computing the reasoning ability by analyzing causal relations between entities and events comprises (i) receiving the textual data pertaining to the user, (ii) identifying a plurality of entities and a plurality of events from the textual data by a Stanford Named Entity Recognizer Tool (iii) generating a plurality of Lexico-syntactic patterns based on the plurality of entities and the plurality of events using regular expressions. (iv) obtaining a plurality of regular expressions by generating a regular expression corresponding to each of the plurality of Lexico-syntactic patterns (v) identifying a plurality of causal relations between each of the plurality events and each of the plurality of entities based on the plurality of regular expressions and (vi) computing the reasoning ability of the user based on the plurality of plurality of causal relations.
[0033] In an embodiment, the plurality of Lexico-syntactic patterns includes a simple causative verb reflecting a causal action, a plurality of phrasal verbs, a plurality of noun-preposition pairs, a plurality of passive causative verbs and a plurality of simple prepositions.
[0034] In an embodiment, the communication skill is computed based on the textual data. The textual data is pre-processed to obtain clean textual data by removing irrelevant data. Further, a plurality of linguistic features are computed from the clean textual data. The plurality of linguistic features includes a plurality of dependency relationship values, a text coherence value and a lexical diversity

value. Further, a plurality of psychological features are simultaneously computed from the clean textual data based on a psycholinguistic analysis. Further, a concatenated feature vector is computed based on the plurality of linguistic features and plurality of psychological features by a first Fully Connected Neural Network (FCNN). A contextual embedding is simultaneously computed based on the clean textual data by a Bidirectional Encoding Representations from Transformers (BERT). Finally, the communication skill of the user is evaluated based on the concatenated feature vector and the contextual embedding by a second FCNN.
[0035] At step 206 of the method 200, the one or more hardware processors 102 are configured by the programmed instructions to compute a unified text based competency value as explained in conjunction with steps of flow diagram depicted in FIG. 3. The computation is based on the plurality of textual features and a corresponding predetermined weight associated with each of the plurality of textual features using a linear regression model. In an embodiment, the linear regression model for computing the unified text based competency value is given in equation
(4).
UT(di) = w1 * stand - out factor + w2 * drive for result value + w3 *
reasoning ability value + w4 * communication skill value +
ε (4)
where w1, w2, w3 and w4 are the weights corresponding to the stand-out factor, the drive for result value, the reasoning ability value and the communication skill value. ε is a regression constant whose value is either empirically determined or can be learned from the training dataset.
[0036] The FIG. 3 is a functional block diagram (300) for unified text based competency value computation for the processor implemented method for machine learning based role fitment assessment implemented by the system of FIG. 1, in accordance with some embodiments of the present disclosure. Now referring to FIG.3, the functional block diagram includes a stand-out factor estimator 302, a drive for result estimator 304, a reasoning ability estimator 306, a communication skill analyzer 308 and a regression model 310. The stand-out factor estimator 302 computes the stand-out factor based on the entropy and the conceptual uniqueness

value. The drive for result estimator 306 computes the drive for result value based on the achiever score, the powerfulness score, the driving factor score, the affect score and the insight score. The reasoning ability estimator 306 computes the reasoning ability value by analyzing causal relations between entities and events. The communication skill analyzer 308 computes the communication skill value based on the plurality of linguistic features, a plurality of psychological features and a word embedding associated with the textual data. The stand-out factor, the drive for result value, the reasoning ability value and the communication skull value is assigned a corresponding weight to obtain a unified text based vector. Further, the unified text based competency value is computed based on t the unified text based vector by the linear regression model 310.
[0037] Referring back to step 208 of the method 200, the one or more hardware processors 102 are configured by the programmed instructions to simultaneously compute a plurality of audiovisual features from the audiovisual data by a plurality of machine learning models, explained in conjunction with flow diagram depicted in FIG. 4. The plurality of audiovisual features comprise a confidence value and an emotion value, a modulation score, and a speaking quality score. The speaking quality score is computed from the audiovisual data based on a total number of pauses and a time duration of each pause.
[0038] The FIG. 4 is a functional block diagram (400) for audio-visual feature computation for the processor implemented method for machine learning based role fitment assessment implemented by the system of FIG. 1, in accordance with some embodiments of the present disclosure. Now referring to FIG. 4, the functional block diagram includes a segmentation module 402, a confidence detection model 404, a confidence analysis and statistical measures module 406, an emotion detection model 408, an emotion analytics and statistical measures module 410, a score calculation module 412, and a pause duration computation module 414. The audiovisual data is segmented into a plurality of frames by the segmentation module 402. The speaking quality score of the user is computed by the pause duration computation module 414 based on the duration, frequency and position of the paused frames. The plurality of frames are further given as input to the

confidence detection model 404 which predicts confidence for the plurality of frames. The predicted confidence is given as input to the confidence analysis and statistical measures module 406, which together compute the confidence value of the user. The emotion detection model 408 predicts emotion for the plurality of frames and the predicted confidence is given as input to the emotion analysis and statistical measures module 410, which compute the emotion value of the user. The computed emotion value and the computed confidence value are provided as input to the score calculator module 412, which compute the modulation score of the user based on the variation between the computed emotion value and the computed confidence value.
[0039] Referring back to method 200, at step 210 , the one or more hardware processors 102 are configured by the programmed instructions to simultaneously compute a plurality of personality dimensions based on the survey data, wherein the survey data is obtained based on a questionnaire.
[0040] In an embodiment, computing the plurality of personality dimensions based on the survey data includes receiving the survey data. The survey data is obtained based on a questionnaire that uses Likert’s scale measures. Further, a plurality of survey features are computed by a plurality techniques including standard scoring methods and a reverse coding. The plurality of survey features includes a plurality of personality traits an achievement orientation, a self-efficacy, a technical knowledge, an initiative tendency, a customer service orientation, an interpersonal communication, an information seeking tendency, an analytical thinking, a customer responsiveness, an impact and influence analysis, a resilience and a sales performance. The plurality of personality traits includes openness, consciousness, extraversion, an agreeableness, a neuroticism.
[0041] In an embodiment, the survey data is measured with a set of statements (items) and the response was obtained on a 5 point Likert scale (For example, Strongly Agree, Agree, Neutral, Disagree, Strongly Disagree) and scores for each sales competency dimension was arrived at. Finally, plurality of personality dimensions is computed based on the plurality of survey features by a Principal Component Analysis (PCA). The plurality of predefined number of

personality dimensions includes an influence creation, a customer centricity, an emotional balance, an analytical thinking and a drive for action.
[0042] At step 212 of the method 200, the one or more hardware processors 102 are configured by the programmed instructions to compute a role fitment value of the user based on the unified text based competency score, the plurality of audiovisual features and the plurality of dimensions of competency by a random forest classification.
[0043] FIG. 5 illustrates an overview of an example architecture (500) for the processor implemented method for machine learning based role fitment assessment implemented by the system of FIG. 1, in accordance with some embodiments of the present disclosure. Now referring to FIG. 3, the architecture includes a standard scoring and reverse scoring module 502, a PCA module 504, a linguistic processing unit 506, the regression model 508, a machine learning unit 510 and a random forest classifier 512. The survey data is given as input to the standard scoring and reverse scoring module 502 which generates the plurality of survey features. The plurality of survey features are given as input to the PCA module 504 to compute the plurality of dimensions of competency including the influence creation, the customer centricity, the emotional balance, the analytical thinking and the drive for action. The textual data is given as input to the linguistic processing unit 506 which generates the plurality of textual features including the stand-out factor, the drive for result value, the reasoning ability value and the communication skill value. Further, the plurality of textual features are given as input to the regression model 508 which computes the unified text based competency value. Similarly, the audio-visual data is given as input to the machine learning unit 510 which generates the plurality of audiovisual features including the confidence value, the emotion value, the modulation score and the speaking quality score. The plurality of audio-visual features, the unified text based competency and the plurality of dimensions of competency are given as input to the random forest classifier 512 which generates the role fitment value associated with the user.
[0044] 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.
[0045] The embodiments of present disclosure herein address the unresolved problem of machine learning based role fitment assessment. The present disclosure provides an efficient method for assessing the role fitment or competency of a user. Here, the plurality of textual features and the audio-visual features are computed in a unique way which increases the efficiency of the system. For example, the stand-out factor is computed based on an entropy and a concept uniqueness. The reasoning ability value is computed based on the causal relation between the plurality of entities and the plurality of events associated with the textual data. The drive for result value is computed based on the plurality of stylometric feature.
[0046] 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 modules 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, GPUs and edge computing devices.

[0047] 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 modules described herein may be implemented in other modules or combinations of other modules. 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. 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 and spirit 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. 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. 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.
[0048] 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 (200), the method comprising:
receiving (202), by one or more hardware processors, an assessment data pertaining to a user, wherein the assessment data comprises a textual data, a survey data and an audiovisual data, wherein the user is associated with a competency profile;
computing (204), by the one or more hardware processors, a plurality of textual features based on the textual data by a linguistic text analysis technique, wherein the plurality of textual features comprises a stand-out factor, a drive for result value, a reasoning ability value and a communication skill value;
computing (206), by the one or more hardware processors, a unified text based competency value based on the plurality of textual features and a corresponding predetermined weight associated with each of the plurality of textual features by a linear regression model;
simultaneously computing (208), by the one or more hardware processors, a plurality of audiovisual features from the audiovisual data by a machine learning technique, wherein the plurality of audiovisual features comprises a confidence value, an emotion value, a modulation score and a speaking quality score;
simultaneously computing (210), by the one or more hardware processors, a plurality of dimensions of competency based on the survey data, wherein the survey data is obtained based on the competency profile of the user; and
computing (212), by the one or more hardware processors, a role fitment value of the user based on the unified text based competency score, the plurality of audiovisual features and the plurality of survey features by a random forest classification.

2. The method as claimed in claim 1, wherein the method of computing the stand-out factor based on an entropy and a concept uniqueness associated with the textual data comprising:
receiving the textual data pertaining to the user; computing an entropy based on the textual data by:
generating a plurality of unique bag-of-words based on the textual data by counting number of times each of the plurality of unique bag-of-word occurs in the textual data, wherein each of the plurality of unique bag-of-words is associated with a corresponding weight;
obtaining a plurality of smoothed bag-of-words based on the plurality of unique bag-of-words using a linear interpolation smoothing, wherein the linear interpolation smoothing removes zero probability problem when any one of the corresponding weights is zero; and
computing the entropy based on each of the plurality of smoothed bag-of-words and the corresponding weight; simultaneously computing a plurality of key-phrases from the textual data with a corresponding confidence score by a Rapid Automatic Keyword Extraction (RAKE) algorithm;
obtaining a plurality of confident key-phrases from the plurality of key-phrases based on a predetermined confidence threshold by RAKE, wherein the plurality of key-phrases with the confidence score greater than the predetermined confidence threshold are selected;
computing a corresponding frequency of occurrence of each of the plurality of confident key-phrases by counting a number of times each of the plurality of confident key-phrase occurs in the textual data;
obtaining a plurality of unique unigrams by sorting the plurality of confident key-phrases in descending order based on the corresponding frequency of occurrence;

obtaining a plurality of noun unigrams based on the plurality of unique unigrams from a WordNet database;
obtaining a plurality of unigram based multiword phrases from the plurality of confident key-phrases based on the plurality of noun unigrams;
generating a candidate elaboration list based on the plurality of unigram based multiword phrases by selecting the plurality of confident key-phrases containing at least one noun unigram from the plurality of noun unigrams;
obtaining a plurality of root words based on the candidate elaboration list by generating a root form for each word in the candidate elaboration list using Porter’s stemming algorithm;
obtaining a plurality of concepts by arranging each of the plurality of root words in alphabetical order;
computing a concept uniqueness value for each of the plurality of concepts by computing an inverted ratio between a number of documents containing the concept and a total number of document; and
computing the stand-out-factor based on the entropy, the concept uniqueness value and a number of concepts, wherein the number of concepts is a total number of the plurality of concepts.
3. The method as claimed in claim 1, wherein the method of computing the drive for result value based on a plurality of stylometric feature values comprising:
receiving the textual data pertaining to the user;
obtaining the plurality of stylometric feature values from the textual data using a Linguistic Information and Word Count (LIWC) tool, wherein the plurality of stylometric feature values comprises an achiever score, a powerfulness score, a driving factor score, an affect score and an insight score; and

computing the drive for result value based on the plurality of stylometric features by computing a weighted average of each of the plurality of stylometric feature scores.
4. The method as claimed in claim 1, wherein the method of computing the
reasoning ability value based on causal relation between a plurality of
entities and a plurality of events associated with the textual data comprising:
receiving the textual data pertaining to the user;
identifying the plurality of entities and a plurality of events from the textual data by a Stanford Named Entity Recognizer Tool;
generating a plurality of Lexico-syntactic patterns based on the plurality of entities and the plurality of events using regular expressions, wherein the plurality of Lexico-syntactic patterns comprise a simple causative verb reflecting a causal action, a plurality of phrasal verbs, a plurality of noun-preposition pairs, a plurality of passive causative verbs and a plurality of simple prepositions;
obtaining a plurality of regular expressions by generating a regular expression corresponding to each of the plurality of Lexico-syntactic patterns;
identifying a plurality of causal relations between each of the plurality events and each of the plurality of entities based on the plurality of regular expressions; and
computing the reasoning ability value of the user based on the plurality of causal relations.
5. The method as claimed in claim 1, wherein the speaking quality score is computed from the audiovisual data based on a total number of pauses and a time duration of each pause.
6. The method as claimed in claim 1, wherein the method of computing the plurality of dimensions of competency based on the survey data comprising:

receiving the survey data, wherein the survey data is obtained based on a questionnaire;
computing a plurality of survey features from the survey data by a plurality techniques comprising a standard scoring method and a reverse coding, wherein the plurality of survey features comprise a plurality of personality traits, an achievement orientation, a self-efficacy, a technical knowledge, an initiative tendency, a customer service orientation, an interpersonal communication, an information seeking tendency, an analytical thinking, a customer responsiveness, an impact and influence creation, a resilience and a sales performance, wherein the plurality of personality traits comprises an openness, a conscientiousness, an extraversion, an agreeableness and a neuroticism; and
computing the plurality of dimensions of competency based on the plurality of survey features by a Principal Component Analysis (PCA), wherein the plurality of dimensions of competency comprise an influence creation, a customer centricity, an emotional balance, an analytical thinking and a drive for action.
7. The method as claimed in claim 1, wherein the communication skill value is computed based on a plurality of linguistic features, a plurality of psychological features and a word embedding associated with the textual data.
8. A system (100) comprising:
at least one memory (104) storing programmed instructions; one or more Input /Output (I/O) interfaces (112); and one or more hardware processors (102) operatively coupled to the at least one memory (104), wherein the one or more hardware processors (102) are configured by the programmed instructions to:

receive an assessment data pertaining to a user, wherein the assessment data comprises a textual data, a survey data and an audiovisual data, wherein the user is associated with a competency profile;
compute a plurality of textual features based on the textual data by a linguistic text analysis technique, wherein the plurality of textual features comprises a stand-out factor, a drive for result value, a reasoning ability value and a communication skill value;
compute a unified text based competency value based on the plurality of textual features and a corresponding predetermined weight associated with each of the plurality of textual features by a linear regression model;
simultaneously compute a plurality of audiovisual features from the audiovisual data by a machine learning technique, wherein the plurality of audiovisual features comprises a confidence value, an emotion value, a modulation score and a speaking quality score;
simultaneously compute a plurality of dimensions of competency based on the survey data, wherein the survey data is obtained based on the competency profile of the user; and
compute a role fitment value of the user based on the unified text based competency score, the plurality of audiovisual features and the plurality of survey features by a random forest classification.
9. The system of claim 8, wherein the method of computing the stand-out factor based on an entropy and a concept uniqueness associated with the textual data comprising:
receiving the textual data pertaining to the user; computing an entropy based on the textual data by:
generating a plurality of unique bag-of-words based on the textual data by counting number of times each of the plurality of unique bag-of-word occurs in the textual data,

wherein each of the plurality of unique bag-of-words is associated with a corresponding weight;
obtaining a plurality of smoothed bag-of-words based on the plurality of unique bag-of-words using a linear interpolation smoothing, wherein the linear interpolation smoothing removes zero probability problem when any one of the corresponding weights is zero; and
computing the entropy based on each of the plurality of smoothed bag-of-words and the corresponding weight; simultaneously computing a plurality of key-phrases from the textual data with a corresponding confidence score by a Rapid Automatic Keyword Extraction (RAKE) algorithm;
obtaining a plurality of confident key-phrases from the plurality of key-phrases based on a predetermined confidence threshold by RAKE, wherein the plurality of key-phrases with the confidence score greater than the predetermined confidence threshold are selected;
computing a corresponding frequency of occurrence of each of the plurality of confident key-phrases by counting a number of times each of the plurality of confident key-phrase occurs in the textual data;
obtaining a plurality of unique unigrams by sorting the plurality of confident key-phrases in descending order based on the corresponding frequency of occurrence;
obtaining a plurality of noun unigrams based on the plurality of unique unigrams from a WordNet database;
obtaining a plurality of unigram based multiword phrases from the plurality of confident key-phrases based on the plurality of noun unigrams; generating a candidate elaboration list based on the plurality of unigram based multiword phrases by selecting the plurality of confident key-phrases containing at least one noun unigram from the plurality of noun unigrams;

obtaining a plurality of root words based on the candidate elaboration list by generating a root form for each word in the candidate elaboration list using Porter’s stemming algorithm;
obtaining a plurality of concepts by arranging each of the plurality of root words in alphabetical order;
computing a concept uniqueness value for each of the plurality of concepts by computing an inverted ratio between a number of documents containing the concept and a total number of document; and
computing the stand-out-factor based on the entropy, the concept uniqueness value and a number of concepts, wherein the number of concepts is a total number of the plurality of concepts.
10. The system of claim 8, wherein the method of computing the drive for result
value based on a plurality of stylometric feature values comprising:
receiving the textual data pertaining to the user;
obtaining the plurality of stylometric feature values from the textual data using a Linguistic Information and Word Count (LIWC) tool, wherein the plurality of stylometric feature values comprises an achiever score, a powerfulness score, a driving factor score, an affect score and an insight score; and
computing the drive for result value based on the plurality of stylometric features by computing a weighted average of each of the plurality of stylometric feature scores.
11. The system of claim 8, wherein the method of computing the reasoning
ability value based on causal relation between a plurality of entities and a
plurality of events associated with the textual data comprising:
receiving the textual data pertaining to the user;
identifying the plurality of entities and a plurality of events from the textual data by a Stanford Named Entity Recognizer Tool;

generating a plurality of Lexico-syntactic patterns based on the plurality of entities and the plurality of events using regular expressions, wherein the plurality of Lexico-syntactic patterns comprise a simple causative verb reflecting a causal action, a plurality of phrasal verbs, a plurality of noun-preposition pairs, a plurality of passive causative verbs and a plurality of simple prepositions;
obtaining a plurality of regular expressions by generating a regular expression corresponding to each of the plurality of Lexico-syntactic patterns;
identifying a plurality of causal relations between each of the plurality events and each of the plurality of entities based on the plurality of regular expressions; and
computing the reasoning ability value of the user based on the plurality of causal relations.
12. The system of claim 8, wherein the speaking quality score is computed from the audiovisual data based on a total number of pauses and a time duration of each pause.
13. The system of claim 8, wherein the method of computing the plurality of dimensions of competency based on the survey data comprising:
receiving the survey data, wherein the survey data is obtained based on a questionnaire;
computing a plurality of survey features from the survey data by a plurality techniques comprising a standard scoring method and a reverse coding, wherein the plurality of survey features comprise a plurality of personality traits, an achievement orientation, a self-efficacy, a technical knowledge, an initiative tendency, a customer service orientation, an interpersonal communication, an information seeking tendency, an analytical thinking, a customer responsiveness, an impact and influence creation, a resilience and a sales performance, wherein the plurality of

personality traits comprises an openness, a conscientiousness, an extraversion, an agreeableness and a neuroticism; and
computing the plurality of dimensions of competency based on the plurality of survey features by a Principal Component Analysis (PCA), wherein the plurality of dimensions of competency comprise an influence creation, a customer centricity, an emotional balance, an analytical thinking and a drive for action.
14. The system of claim 8, wherein the communication skill value is computed based on a plurality of linguistic features, a plurality of psychological features and a word embedding associated with the textual data.