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 USER REVIEW BASED COGNITIVE PRODUCT RECOMMENDATION
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 Natural language Processing (NLP) and, more particular, to a method and system for user review based cognitive product recommendation.
BACKGROUND
[002] Product recommendation is a part of any e-commerce personalization strategy used to dynamically populate products of user’s interest for providing a personalized shopping experience. The recommendation is made based on customer attributes, browsing behavior and situational context. With the growing amount of information on the internet and with a significant rise in the number of users, it is becoming important for companies to search, map and provide them with the relevant chunk of information according to their preferences and taste.
[003] Conventional methods for product recommendation are using user’s historical buying behavior to give entity recommendation. There are some user reviews based methods wherein the polarity scores are biased more towards negative sentiments. The polarity scores indicates whether a sentiment associated with the user review is positive or negative. Hence there is a challenge in implementing a system which understands the subjective and qualitative nature of unrelated user search query in the realm of recommendation without any polarity bias.
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 user review based cognitive product recommendation is provided. The method includes receiving, by one or more hardware processors, a user query pertaining to a product, wherein the user query comprises a plurality of subjective words indicating a sentiment. Further, the method includes identifying via the one or more hardware processors, a plurality of noun entities by parsing the

user query using a Natural Language Processing (NLP) technique, wherein the noun entities includes a plurality of product categories and a plurality of product features. Furthermore, the method includes simultaneously identifying via the one or more hardware processors, a plurality of subjective criteria based on a plurality of subjective words associated with the user query using the NLP technique. Furthermore, the method includes extracting via the one or more hardware processors, a first set of tuples from a polarity metadata repository corresponding to the product based on the plurality of product categories, wherein the polarity metadata repository is generated using a normalized polarity computation technique. Furthermore, the method includes extracting via the one or more hardware processors, a second set of tuples from the first set of tuples associated with the polarity metadata repository based on the plurality of product features. Furthermore, the method includes obtaining via the one or more hardware processors, a plurality of sorted tuples by sorting the second set of tuples in one of, an ascending order and a descending order based on one of a subjective criteria out of the plurality of subjective criteria. Finally, the method includes recommending via the one or more hardware processors, a plurality of products corresponding to the user query based on the plurality of sorted tuples, wherein the plurality of products pertaining to a first predefined number of sorted tuples.
[005] In another aspect, a system for user review based cognitive product recommendation 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 a user query pertaining to a product, wherein the user query comprises a plurality of subjective words indicating a sentiment. Further, the one or more hardware processors are configured by the programmed instructions to identify a plurality of noun entities by parsing the user query using a Natural Language Processing (NLP) technique, wherein the noun entities includes a plurality of product categories and a plurality of product features. Furthermore, the one or more hardware processors are configured by the programmed instructions to

simultaneously identify a plurality of subjective criteria based on a plurality of subjective words associated with the user query using the NLP technique. Furthermore, the one or more hardware processors are configured by the programmed instructions to extract a first set of tuples from a polarity metadata repository corresponding to the product based on the plurality of product categories, wherein the polarity metadata repository is generated using a normalized polarity computation technique. Furthermore, the one or more hardware processors are configured by the programmed instructions to extract second set of tuples from the first set of tuples associated with the polarity metadata repository based on the plurality of product features. Furthermore, the one or more hardware processors are configured by the programmed instructions to obtain a plurality of sorted tuples by sorting the second set of tuples in one of, an ascending order and a descending order based on one of a subjective criteria out of the plurality of subjective criteria. Finally, the one or more hardware processors are configured by the programmed instructions to recommend a plurality of products corresponding to the user query based on the plurality of sorted tuples, wherein the plurality of products pertaining to a first predefined number of sorted tuples.
[006] In yet another aspect, a computer program product including a non-transitory computer-readable medium having embodied therein a computer program for user review based cognitive product recommendation is provided. The computer readable program, when executed on a computing device, causes the computing device to receive a user query pertaining to a product, wherein the user query comprises a plurality of subjective words indicating a sentiment. Further, the computer readable program, when executed on a computing device, causes the computing device to identify a plurality of noun entities by parsing the user query using a Natural Language Processing (NLP) technique, wherein the noun entities includes a plurality of product categories and a plurality of product features. Furthermore, the computer readable program, when executed on a computing device, causes the computing device to simultaneously identify a plurality of subjective criteria based on a plurality of subjective words associated with the user query using the NLP technique. Furthermore, the computer readable program, when

executed on a computing device, causes the computing device to extract a first set of tuples from a polarity metadata repository corresponding to the product based on the plurality of product categories, wherein the polarity metadata repository is generated using a normalized polarity computation technique. Furthermore, the computer readable program, when executed on a computing device, causes the computing device to extract second set of tuples from the first set of tuples associated with the polarity metadata repository based on the plurality of product features. Furthermore, the computer readable program, when executed on a computing device, causes the computing device to obtain a plurality of sorted tuples by sorting the second set of tuples in one of, an ascending order and a descending order based on one of a subjective criteria out of the plurality of subjective criteria. Finally, the computer readable program, when executed on a computing device, causes the computing device to recommend a plurality of products corresponding to the user query based on the plurality of sorted tuples, wherein the plurality of products pertaining to a first predefined number of sorted tuples.
[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 user review based cognitive product recommendation, in accordance with some embodiments of the present disclosure.
[0010] FIGS. 2A and 2B are exemplary flow diagrams illustrating a method for user review based cognitive product recommendation, implemented by the system of FIG. 1, in accordance with some embodiments of the present disclosure.
[0011] FIG. 3 illustrates a functional block diagram for generating the polarity metadata repository for the processor implemented method for user review

based cognitive product recommendation implemented by the system of FIG. 1, in accordance with some embodiments of the present disclosure.
[0012] FIG. 4 is an example overall architecture for the processor implemented method for user review based cognitive product recommendation implemented by the system of FIG. 1, in accordance with some embodiments of the present disclosure.
DETAILED DESCRIPTION OF EMBODIMENTS [0013] 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. [0014] Embodiments herein provide a method and system for user review based cognitive product recommendation for recommending a best product to a user without any polarity bias. Initially, the system receives a user query pertaining to a product. The user query includes a plurality of subjective words indicating a sentiment of the user. Further, a plurality of noun entities is identified by parsing the user query using Natural Language Processing (NLP) technique. The noun entities include a plurality of product category and a plurality of product features. Simultaneously, a plurality of subjective criteria are identified based on the plurality of subjective words associated with the user query using NLP. Further, a first set of tuples are extracted from a polarity metadata repository based on the plurality of product category. The polarity metadata repository is generated based on a normalized polarity computation technique. After extracting the first set of tuples, a second set of tuples are extracted from the first set of tuples associated with the polarity metadata repository based on the plurality of product features. Further, a plurality of sorted tuples is obtained by sorting the second set of tuples in one of, an ascending order and a descending order based on the subjective criteria. The

sorting is performed in ascending order when the subjective criteria indicate a positive sentiment. Finally, a plurality of top n products is recommended based on the plurality of sorted tuples.
[0015] Referring now to the drawings, and more particularly to FIGS. 1 through 3, 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.
[0016] FIG. 1 is a functional block diagram of a system 100 for user review based cognitive product recommendation, 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.
[0017] 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.
[0018] 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.

[0019] 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.
[0020] 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.
[0021] The plurality of modules 106 include programs or coded instructions that supplement applications or functions performed by the system 100 for user review based cognitive product recommendation. 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 user review based cognitive product recommendation. In an embodiment, plurality of modules 106 includes an NLP module (not shown in FIG. 1), a tuple extraction module (not shown in FIG. 1), a sorting module (not shown in FIG. 1) and a recommendation module (not shown in FIG. 1).

[0022] 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.
[0023] 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).
[0024] FIGS. 2A and 2B are exemplary flow diagrams illustrating a method 200 for user review based cognitive product recommendation 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. 2A and 2B. 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.
[0025] At step 202 of the method 200, the one or more hardware processors 102 are configured by the programmed instructions to receive a user query pertaining to a product. The user query comprises a plurality of subjective words indicating a sentiment. For example, the user query can be “best mobile with good camera under twenty thousand”. In this user query, the user is expecting a product “mobile phone” with a feature “good camera”. Here, the words indicating the sentiments are “best” and “good”.
[0026] At step 204 of the method 200, the one or more hardware processors 102 are configured by the programmed instructions to identify a plurality of noun entities by parsing the user query using NLP technique. The noun entities include a plurality of product categories and a plurality of product features. For example, in the example query “best mobile with good camera under twenty thousand”, the extracted entities are the “mobile” and “camera”. Here, “mobile” indicates the expected product category and the “camera” indicates the expected feature.
[0027] At step 206 of the method 200, the one or more hardware processors 102 are configured by the programmed instructions to simultaneously identify a plurality of subjective criteria based on the plurality of subjective words associated with the user query using the NLP technique. For example, considering the above user query, the subjective criteria are “good” and “best”.
[0028] At step 208 of the method 200, the one or more hardware processors 102 are configured by the programmed instructions to extract a first set of tuples from a polarity metadata repository based on the plurality of product categories. The polarity metadata repository is generated using a normalized polarity computation technique. For example, the first set of tuples corresponding to the user query “best mobile with good camera under twenty thousand” is given as {'Product': 'mobile', 'Filter': {'camera': 'best'}} and {'Product': 'mobile', 'Price': [('<=', 20000)], 'Filter': {'camera': 'best'}}.

[0029] At step 210 of the method 200, the one or more hardware processors 102 are configured by the programmed instructions to extract a second set of tuples from the first set of tuples associated with the polarity metadata repository based on the plurality of product features. For example, the second set of tuples corresponding to the user query “best mobile with good camera under twenty thousand” is given in Table I.
Table I

Product Entity Weight Top Top All key
ID Positive Key phrase Negative key phrase phrases
love this
camera,
great [['love this
Camera',
'Camera',
0.7221],
['great
Camera',
'Camera',
0.7165], ['nice
Camera',
'Camera',
0.5596],
['smart
Camera',
'Camera',
0.5377], ['new
Camera',
'Camera',
0.1364], ['took
the Camera',
p4 Camera 4.2 camera Nil 'Camera', 0.0]]

[0030] At step 212 of the method 200, the one or more hardware processors 102 are configured by the programmed instructions to obtain a plurality of sorted tuples by sorting the second set of tuples in one of, an ascending order and a descending order based on one of a subjective criteria out of the plurality of subjective criteria. The sorting is performed in ascending order when the subjective criteria indicate a positive sentiment. While performing sorting, the metadata is sorted on the feature specified in the use query. For example, the metadata is sorted based on camera. In an embodiment, when there is no feature specified in user query, for example, “Best mobile under 20000”, the metadata is sorted on all significant features for a product.
[0031] At step 214 of the method 200, the one or more hardware processors 102 are configured by the programmed instructions to recommend a plurality of products corresponding to the user query based on the plurality of sorted tuples, wherein the plurality of products pertaining to a first predefined number of sorted tuples.
[0032] In an embodiment, the method of generating the polarity metadata repository is explained with reference to FIG. 3. FIG. 3 illustrates a functional block diagram for generating the polarity metadata repository for the processor implemented method for user review based cognitive product recommendation 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 sentence pattern identification module 302, the spaCy model 304, a significant entities identification module 306, a clustering module 308, a bias neutralization engine 310, a ranking module 312, a selection module 314 and a polarity metadata repository generation module 316.
[0033] Initially a plurality of user reviews pertaining to a product are received by the sentence pattern identification module 302 and a plurality of review key-phrases are identified based on the plurality of user reviews. For example, the plurality of review key phrased and the corresponding sentence patterns is given in Table II.
Table II

S.No TextData Sentence Pattern
1 absolutely LOVE this product.absolutely love its convenience. Adv+verb+det+noun
2 Love the shine, love the food verb+det+noun
3 easy to use, easy to apply Adj+part+verb
4 Value for Money Noun+ADP+Noun
5 Very comfy shirt! Adv+Adj+noun
6 looks beautiful..feels beautiful verb+Adj
7 Must buy Adv+Verb
8 Cool product! Adj+Noun
9 Very comfortable Adv+Adj
[0034] After identifying the plurality of review key-phrases, a plurality of entities are identified based on the plurality of review key-phrases by parsing each of the plurality of review key-phrases by the spaCy model 304. After identifying the plurality of entities, a plurality of significant entities are obtained based on a frequency of occurrence corresponding to each of the plurality of entities by the significant entities identification module 306. The frequency of occurrence corresponding to each of the plurality of entities is computed based on a cumulative sum of key phrases mentioning each entity across the plurality of reviews. The plurality of entities with the frequency of occurrence greater than a predefined threshold is selected as significant entity. After selecting the plurality of significant entities, a set of review key-phrases (a cluster of review key-phrases) corresponding to each of the plurality of significant entities are selected based on a comparison between each of the plurality of review key-phrases and each of the plurality of significant entities by the clustering module 308. This provides a group/cluster C of review key-phrases corresponding to each of the plurality of significant entities E and all generic key phrases which do not contain any entity reference but represent user sentiments. For example, the user reviews like “value for money”, ”must buy”, “definitely recommended” are grouped under cluster called generic

hence total number of clusters would be denoted by equation (1). Here Ci is the
cluster formed for ith entity Ei. Total number of clusters Cji is equal to a number
of entities Ei ranging from [1…n added with one cluster for generic sentiment key phrases.

[0035] After obtaining the set of review key-phrases corresponding to each of the plurality of significant entities, a polarity score is computed for each review key-phrase from the set of (cluster of) review key-phrases by a polarity computation tool. In an embodiment, the polarity score is computed by Valence Aware Dictionary sEntiment Reasoner (VADER). The plurality of clusters cluster 1 to cluster n of the clustering module 308 are given as input to a bias neutralization 310. Here, in each cluster KP1 indicated the polarity value for review key-phrase 1 corresponding to the cluster. The bias neutralization engine 310 neutralizes the bias towards positive review key-phrases and the negative key-phrases using a normalization technique. The bias neutralization engine 310 generates a cumulative normalized polarity score corresponding to each of the plurality of clusters using a modified sigmoid function. After computing the cumulative normalized polarity score for each cluster of review key-phrases, the set of review key-phrases corresponding to each of the plurality of significant entities are ranked by a ranking module 312 based on the corresponding cumulative normalized polarity score. Here, the ranking is done in descending order. Further, the set of review key-phrases with a ranking greater than a predefined ranking threshold is selected by the selection module 314. Finally, the polarity metadata repository for the product is generated based on the set of review key-phrases corresponding to each of the plurality of significant entities by a polarity metadata repository generation module 316. The polarity metadata repository of the product includes a plurality of significant entity names, the corresponding normalized polarity score, the plurality of review key-phrases, a plurality of positive key phrases and a plurality of negative key phrases. For example, a sample plurality metadata repository looks as shown in Table III.

Table III

Product Top Positive Top Negative
id Entity weight Key phrases Key phrases All Key phrases
very good [['very good
quality', 'quality',
0.7519], ['brilliant
quality', 'quality',
0.6959,['not as
Quality quality, powerful',
brilliant not as 'generic', -
p4 4.4 quality powerful 0.3754]]
Camera love this
camera,
great [['love this
Camera',
'Camera',
0.7221], ['great
Camera',
'Camera',
0.7165], ['nice
Camera',
'Camera',
0.5596], ['smart
Camera',
'Camera',
0.5377], ['new
Camera',
'Camera',
0.1364], ['took
the Camera',
p4 4.2 camera 'Camera', 0.0]]

not like
performance,
not like the
Camera bad [['great
performance',
'performance',
0.7165], ['not like
performance',
'performance', -
0.3242], ['not like
the Camera bad
camera bad
performance',
Perfor great camera bad 'performance', -
P5 Mance 2.7 performance performance 0.8251]]
[0036] In an embodiment, the cumulative normalized polarity score for each group of review key-phrases is denoted by equation (2). Here, is the cluster
formed for each entity ,pi is the positive polarity for ith key phrase , Ni is the
negative polarity for key phrase. Hence Cumulative polarity for cluster for
ith entity is the summation of positive and negative polarities for all the key phrases where key phrases ranges from [1…n].

[0037] In an embodiment, the bias neutralization engine for computing the cumulative normalized polarity score is a mathematical deduction technique which eliminates the offset using the modified sigmoid function. Here, the polarity for the sentiment (subjective criteria) is achieved with the help of rule based method, wherein lexicon containing weights for each review key-phrase is used as a reference. The polarity score mentioned below is taken from weights associated with lexicon and plotted with the help of modified sigmoid such that it would lie between -1 to +1. The sigmoid function has been modified for different set of values in such a manner that while removing negative bias, an incremental positive bias is

not getting added for positive sentiments. The modified sigmoid function for a plurality of polarity scores are given in equation (3) to (8). The normalization function is given in equation (3).

polarity_score = sigmoid (score + ( 1/ score2 )), if score >=
1 …………………………(5)
polarity_score = sigmoid (score - (1⁄score2)), if score <=
-1 ……..………...…..…..(6)
polarity_score = sigmoid(score - score2), if -1< score < 0 …………………..…….(7) polarity_score = sigmoid(score + score2), if 0 < score < 1 ………….……………….(8) [0038] FIG. 4 is an example overall architecture (400) for the processor implemented method for user review based cognitive product recommendation implemented by the system of FIG. 1, in accordance with some embodiments of the present disclosure. Now referring to FIG. 4, the user query is received by the NLP module 402 and the NLP module 402 generates the plurality of review key-phrases and the plurality of entities. Simultaneously the NLP module 402 identifies the plurality of subjective criteria based on the plurality of subjective words associated with the user query. The tuple extraction module extracts the first set of tuples from the polarity metadata repository 404 corresponding to the product based on the plurality of product category. The polarity metadata repository is generated based on a normalized polarity computation technique. Further, the tuple extraction module 404 extracts the second set of tuples from the first set of tuples associated with the polarity metadata repository based on the plurality of product features. Further, the sorting module 408 obtains the plurality of sorted tuples by sorting the second set of tuples in one of, an ascending order and a descending order based on the subjective criteria. The sorting is performed in ascending order when the subjective criteria indicate a positive sentiment. The recommendation module 410 recommends the plurality of products based on the plurality of sorted tuples. Here,

the plurality of products pertaining to the first n number of sorted tuples are selected for recommendation.
[0039] In an embodiment, the present disclosure is experimented, and the details of experimentation are explained, and the results are explained below. In an embodiment, the Table IV provides a plurality of input queries and the corresponding recommendation provided by the system 100.
Table IV

Sl. Input user Product recommendation Sorting module
No 1
2 query logic

mobiles with best camera under 10000 ([Product 1, Product 2, product 3], [('Product', 'Mobile'), ('camera', 'best')]) Sorting on entity Camera

Best mobile ([Product 1, Product 2, product Sorting on all
under 20000 3], [('Product', 'Mobile'), significant
('camera',’Display’,’’processor’ entities
'best')]) (significant entities in this example are camera,Display
3 and processor)

Mobiles with ([Product 1, Product 2, product Sorting on
best camera 3], [('Product', 'Mobile'), entity Camera
and display ('camera',’Display’, 'best')]) and display

[0040] The polarity score assignment is biased more towards negative sentiments in the prior arts. For example, as data set increases for extreme sentiments then there is a negative bias that gets added to the state of the art library when it should remain as a neutral bias. However, the present disclosure can neutralize the positive and the negative biases as shown in Table V.
Table V

Sr No Frequency of Occurrence Pattern Key phrases Prior arts Present disclosure
1 1 Excellent worst 2.8 3
2 2 Excellent worst 2.6 2.9
3 3 Excellent worst 2.4 2.9
4 4 Excellent worst 2.2 2.8
5 5 Excellent worst 2.1 2.8
6 10 Excellent worst 1.6 2.5
7 15 Excellent worst 1.3 2.1
[0041] The present disclosure is able to use existing lexicon weights and generate polarities in such a manner that while removing negative bias it should not add incremental positive bias and provides consistent results for all sentiment keywords.
[0042] 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.
[0043] The embodiments of present disclosure herein address the unresolved problem of user review based cognitive product recommendation. The present disclosure removes the bias created using positive and negative sentiments by using the bias neutralization engine implemented using the modified sigmoid function. Here, the polarity metadata repository is created for each product using a unique normalized polarity computation technique which helps in accurate and efficient recommendation of the product.
[0044] 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.
[0045] 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.
[0046] 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), via one or more hardware processors, a user query pertaining to a product, wherein the user query comprises a plurality of subjective words indicating a sentiment;
identifying (204), via the one or more hardware processors, a plurality of noun entities by parsing the user query using a Natural Language Processing (NLP) technique, wherein the noun entities comprises a plurality of product categories and a plurality of product features;
simultaneously identifying (206), via the one or more hardware processors, a plurality of subjective criteria based on a plurality of subjective words associated with the user query using the NLP technique;
extracting (208), via the one or more hardware processors, a first set of tuples from a polarity metadata repository corresponding to the product based on the plurality of product categories, wherein the polarity metadata repository is generated using a normalized polarity computation technique;
extracting (210), via the one or more hardware processors, a second set of tuples from the first set of tuples associated with the polarity metadata repository based on the plurality of product features;
obtaining (212), via the one or more hardware processors, a plurality of sorted tuples by sorting the second set of tuples in one of, an ascending order and a descending order based on one of a subjective criteria out of the plurality of subjective criteria; and
recommending (214), via the one or more hardware processors, a plurality of products corresponding to the user query based on the plurality of sorted tuples, wherein the plurality of products pertaining to a first predefined number of sorted tuples.
2. The method as claimed on claim 1, wherein the method of generating the
polarity metadata repository comprising:

receiving a plurality of user reviews pertaining to a product;
identifying a plurality of review key-phrases based on the plurality of user reviews by a plurality of predefined sentence patterns;
identifying a plurality of entities based on the plurality of review key-phrases by parsing each of the plurality of review key-phrases by a pre-trained spaCy model;
obtaining a plurality of significant entities based on a frequency of occurrence corresponding to each of the plurality of entities, wherein the frequency of occurrence corresponding to each of the plurality of entities is computed based on a cumulative sum of key phrases mentioning each entity across the plurality of reviews, wherein the plurality of entities with the frequency of occurrence greater than a predefined threshold is selected as significant entity;
generating a set of review key-phrases corresponding to each of the plurality of significant entities based on a comparison between each of the plurality of review key-phrases and each of the plurality of significant entities;
computing a polarity score for each review key-phrase from the set of review key-phrases by a polarity computation tool;
computing a cumulative normalized polarity score for each set of review key-phrases corresponding to each of the plurality of significant entities by a modified sigmoid function based bias neutralization technique;
ranking the set of review key-phrases corresponding to each of the plurality of significant entities based on the corresponding cumulative normalized polarity score, wherein the ranking is done in descending order;
selecting the set of review key-phrases with a ranking greater than a predefined ranking threshold; and
generating the polarity metadata repository for the product based on the set of review key-phrases corresponding to each of the plurality of significant entities.

3. The method as claimed in claim 1, wherein the polarity metadata repository comprises a plurality of significant entity names, the corresponding normalized polarity score, the plurality of review key-phrases, a plurality of positive key phrases and a plurality of negative key phrases.
4. The method as claimed in claim 1, wherein the sorting is performed in ascending order when the subjective criteria indicate a positive sentiment.
5. 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 a user query pertaining to a product, wherein the user query comprises a plurality of subjective words indicating a sentiment;
identify a plurality of noun entities by parsing the user query using a Natural Language Processing (NLP) technique, wherein the noun entities comprises a plurality of product categories and a plurality of product features;
simultaneously identify a plurality of subjective criteria based on a plurality of subjective words associated with the user query using the NLP technique;
extract a first set of tuples from a polarity metadata repository corresponding to the product based on the plurality of product categories, wherein the polarity metadata repository is generated using a normalized polarity computation technique;
extract a second set of tuples from the first set of tuples associated with the polarity metadata repository based on the plurality of product features;

obtain a plurality of sorted tuples by sorting the second set of tuples in one of, an ascending order and a descending order based on one of a subjective criteria out of the plurality of subjective criteria; and
recommend a plurality of products corresponding to the user query based on the plurality of sorted tuples, wherein the plurality of products pertaining to a first predefined number of sorted tuples.
6. The system of claim 5, wherein the method of generating the polarity metadata repository comprising:
receiving a plurality of user reviews pertaining to a product;
identifying a plurality of review key-phrases based on the plurality of user reviews by a plurality of predefined sentence patterns;
identifying a plurality of entities based on the plurality of review key-phrases by parsing each of the plurality of review key-phrases by a pre-trained spaCy model;
obtaining a plurality of significant entities based on a frequency of occurrence corresponding to each of the plurality of entities, wherein the frequency of occurrence corresponding to each of the plurality of entities is computed based on a cumulative sum of key phrases mentioning each entity across the plurality of reviews, wherein the plurality of entities with the frequency of occurrence greater than a predefined threshold is selected as significant entity;
generating a set of review key-phrases corresponding to each of the plurality of significant entities based on a comparison between each of the plurality of review key-phrases and each of the plurality of significant entities;
computing a polarity score for each review key-phrase from the set of review key-phrases by a polarity computation tool;
computing a cumulative normalized polarity score for each set of review key-phrases corresponding to each of the plurality of significant entities by a modified sigmoid function based bias neutralization technique;

ranking the set of review key-phrases corresponding to each of the plurality of significant entities based on the corresponding cumulative normalized polarity score, wherein the ranking is done in descending order;
selecting the set of review key-phrases with a ranking greater than a predefined ranking threshold; and
generating the polarity metadata repository for the product based on the set of review key-phrases corresponding to each of the plurality of significant entities.
7. The system of claim 5, wherein the polarity metadata repository comprises a plurality of significant entity names, the corresponding normalized polarity score, the plurality of review key-phrases, a plurality of positive key phrases and a plurality of negative key phrases.
8. The system of claim 5, wherein the sorting is performed in ascending order when the subjective criteria indicate a positive sentiment.