Method and system for mapping technical documents for a user query and generating a corpus

Abstract

The present disclosure relates to field of information retrieval that discloses method of mapping technical documents for a user query. A computing system receives mapping input comprising input data in at least one data form of plurality of data forms and type of mapping to be performed. Extracts textual data from input data received in at least one data form of plurality of data forms. Determines using trained Artificial Intelligence (Al) models, first set of candidate phrases from textual data based on type of mapping and technical domain using pre- generated corpus. Extracts contextually intelligent information supplementing textual data based on type of mapping and technical domain of textual data using pre-generated corpus, textual data, external data accessed from external data sources. Generates second set of candidate phrases based on correlation of extracted contextually intelligent information, first set of candidate phrases. Generates report comprising relevant technical documents conformant with input data.

Description

METHOD AND SYSTEM FOR MAPPING TECHNICAL DOCUMENTS FOR A USER QUERY AND GENERATING A CORPUSTECHNICAL FIELD[1] The present disclosure relates, in general, to information retrieval. Particularly, the present disclosure relates to a method and system for mapping technical documents for a user query and generating a corpus.BACKGROUND[2] Users such as physicians, researchers, product owners and product developers often analyze technical literature to learn about the current developments. Searching and mapping of products / idea with technical literature or vice versa is performed to understand the uniqueness of the products or the idea. Some users search for a product which is similar to their idea before investing time and money in developing their idea which saves the cost if the idea is not unique. Also, some users search for products related to their technical literature such as a Patent literature to be aware if the patent is infringed by a third party. There can be numerous instances where the users would like to search or map product with a patent literature. Users such as IP professionals, inventors, users use numerous platforms available on the internet to search for technical documents related to their products.[3] The existing platforms available to search the technical documents perform search based on the keywords provided by the user. However, the user may not be able to search for each keyword present in the document manually. Moreover, technical information related to a product may be presented in various data formats such as a text document, image, videos, audio recordings etc. It is a very tedious and time-consuming process for the users to look into each data format manually to understand the technical details, shortlist the keywords that would be suitable for search and then perform a search to find documents which are similar to the technical information. Further, the existing platforms fail to identify whether there are any technical documents such as patents, scholarly articles etc., for a particular product in the market and / or an idea. Also, the existing platforms fail to identify whether there are any particular products in the market and / or an idea for technical documents such as patents, scholarly articles etc. This type of product is used to technical document mapping and vice versa is not provided by the existing platforms, which may limit the owner of the technical document or the product restrict the third party from using the technical literature without theauthorization of the owner. Therefore, there is a need for an improved method and system that can help with efficient mapping of the information to technical documents.[4] The information disclosed in this background of the disclosure section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art.SUMMARY[5] Disclosed herein is a method of mapping technical documents for a user query. The method comprises receiving, by a computing system, a mapping input comprising input data in at least one data form of a plurality of data forms and a type of mapping to be performed. Further, the method comprises extracting textual data from the input data received in the at least one data form of the plurality of data forms using one or more text extraction techniques. Thereafter, the method comprises determining using one or more trained Artificial Intelligence (Al) models, a first set of candidate phrases from the textual data based on the type of mapping and a technical domain of the textual data using a pre-generated corpus. Furthermore, the method comprises extracting using the one or more trained Al models, contextually intelligent information supplementing the textual data based on the type of mapping and the technical domain of the textual data using at least one of the pre-generated corpus, the textual data, and external data accessed from one or more external data sources associated with the computing system. Thereafter, the method comprises generating a second set of candidate phrases based on correlation of the extracted contextually intelligent information and the first set of candidate phrases. The second set of candidate phrases comprises phrases different from the first set of candidate phrases. Finally, the method comprises generating a report comprising one or more relevant technical documents conformant with the input data by mapping at least one of the first set of candidate phrases and the second set of candidate phrases with one or more technical documents of the technical domain based on the type of mapping and the extracted contextually intelligent information.[6] Further, the present disclosure relates to a computing system for mapping technical documents for a user query. The computing system comprises a processor and a memory. The memory is communicatively coupled to the processor and stores processor-executable instructions, which on execution, cause the processor to receive a mapping input comprisinginput data in at least one data form of a plurality of data forms and a type of mapping to be performed. Further, the processor extracts textual data from the input data received in the at least one data form of the plurality of data forms using one or more text extraction techniques. Thereafter, the processor determines using one or more trained Artificial Intelligence (Al) models, a first set of candidate phrases from the textual data based on the type of mapping and a technical domain of the textual data using a pre-generated corpus. Furthermore, the processor extracts using the one or more trained Al models, contextually intelligent information supplementing the textual data based on the type of mapping and the technical domain of the textual data using at least one of the pre-generated corpus, the textual data, and external data accessed from one or more external data sources associated with the computing system. Thereafter, the processor generates a second set of candidate phrases based on correlation of the extracted contextually intelligent information and the first set of candidate phrases. The second set of candidate phrases comprises phrases different from the first set of candidate phrases. Finally, the processor generates a report comprising one or more relevant technical documents conformant with the input data by mapping at least one of the first set of candidate phrases and the second set of candidate phrases with one or more technical documents of the technical domain based on the type of mapping and the extracted contextually intelligent information.[7] Disclosed herein is a method of generating a corpus. The method comprises receiving, by a computing system, technical data related to a technical domain and corresponding ontology data in at least one data form among a plurality of data forms, from one or more data sources. Further, the method comprises extracting textual data from the technical data and the corresponding ontology data using one or more text extraction techniques. Thereafter, the method comprises determining a first set of technical corpus key phrases related to the technical domain from the extracted textual data using a trained Artificial Intelligence (Al) model. Furthermore, the method comprises determining a second set of technical corpus key phrases related to the technical domain from the extracted textual data based on a user input. The second set of technical corpus key phrases are different from the first set of technical corpus key phrases. Thereafter, the method comprises generating a domain specific knowledge graph indicating a relationship between the first set of technical corpus key phrases and the second set of technical corpus key phrases related to the technical domain. Finally, the method comprises generating a corpus comprising technical corpus key phrases of the technical domainusing the first set of technical corpus key phrases, the second set of technical corpus key phrases and the corresponding domain specific knowledge graph.[8] Further, the present disclosure relates to a computing system generating a corpus. The computing system comprises a processor and a memory. The memory is communicatively coupled to the processor and stores processor-executable instructions, which on execution, cause the processor to receive technical data related to a technical domain and corresponding ontology data in at least one data form among a plurality of data forms, from one or more data sources. Further, the processor extracts textual data from the technical data and the corresponding ontology data using one or more text extraction techniques. Thereafter, the processor determines a first set of technical corpus key phrases related to the technical domain from the extracted textual data using atained Artificial Intelligence (Al) model. Furthermore, the processor determines a second set of technical corpus key phrases related to the technical domain from the extracted textual data based on a user input. The second set of technical corpus key phrases are different from the first set of technical corpus key phrases. Thereafter, the processor generates a domain specific knowledge graph indicating a relationship between the first set of technical corpus key phrases and the second set of technical corpus key phrases related to the technical domain. Finally, the processor generates a corpus comprising technical corpus key phrases of the technical domain using the first set of technical corpus key phrases, the second set of technical corpus key phrases and the corresponding domain specific knowledge graph.[9] The foregoing summary is illustrative only and is not intended to be in any way limiting. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features will become apparent by reference to the drawings and the following detailed description.BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS

[0010] The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, explain the disclosed principles. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The same numbers are used throughout the figures to reference like features and components. Some embodiments of system and / ormethods in accordance with embodiments of the present subject matter are now described, by way of example only, and regarding the accompanying figures, in which:

[0011] FIG. 1A shows an exemplary environment for mapping technical documents for a user query, in accordance with some embodiments of the present disclosure;

[0012] FIG. IB shows an exemplary scenario of input data provided in a plurality of data forms, in accordance with some embodiments of the present disclosure;

[0013] FIG. 1C shows an exemplary document comprising data in one or more data forms, in accordance with some embodiments of the present disclosure;

[0014] FIG. ID shows an exemplary first result list displayed to a user, in accordance with some embodiments of the present disclosure;

[0015] FIG. IE shows an exemplary second result list displayed to a user, in accordance with some embodiments of the present disclosure;

[0016] FIG. IF shows an exemplary environment for generating a corpus, in accordance with some embodiments of the present disclosure;

[0017] FIG. 2 shows a detailed block diagram of the proposed computing system, in accordance with some embodiments of the present disclosure;

[0018] FIG. 3 shows a flowchart illustrating a method of mapping technical documents for a user query, in accordance with some embodiments of the present disclosure;

[0019] FIG. 4 shows a flowchart illustrating a method of generating a corpus for a user query, in accordance with some embodiments of the present disclosure; and

[0020] FIG. 5 illustrates a block diagram of an exemplary computer system for implementing embodiments consistent with the present disclosure.

[0021] It should be appreciated by those skilled in the art that any block diagrams herein represent conceptual views of illustrative systems embodying the principles of the present subject matter. Similarly, it will be appreciated that any flow charts, flow diagrams, state transition diagrams, pseudo code, and the like represent various processes which may besubstantially represented in computer readable medium and executed by a computer or processor, whether such computer or processor is explicitly shown.DETAILED DESCRIPTION

[0022] In the present document, the word “exemplary” is used herein to mean “serving as an example, instance, or illustration.” Any embodiment or implementation of the present subject matter described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0023] While the disclosure is susceptible to various modifications and alternative forms, specific embodiment thereof has been shown by way of example in the drawings and will be described in detail below. It should be understood, however that it is not intended to limit the disclosure to the specific forms disclosed, but on the contrary, the disclosure is to cover all modifications, equivalents, and alternative falling within the scope of the disclosure.

[0024] The terms “comprises”, “comprising”, “includes”, or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a setup, device, or method that comprises a list of components or steps does not include only those components or steps but may include other components or steps not expressly listed or inherent to such setup or device or method. In other words, one or more elements in a system or apparatus proceeded by “comprises. . . a” does not, without more constraints, preclude the existence of other elements or additional elements in the system or method.

[0025] As discussed in the background section, in the existing techniques, efficient mapping of the information to technical documents is not provided. Hence, the present disclosure provides a method and apparatus for mapping technical documents for a user query. The method comprises receiving, by a computing system, a mapping input comprising input data in at least one data form of a plurality of data forms and a type of mapping to be performed. Further, the method comprises extracting textual data from the input data received in the at least one data form of the plurality of data forms using one or more text extraction techniques. Thereafter, the method comprises determining using one or more trained Artificial Intelligence (Al) models, a first set of candidate phrases from the textual data based on the type of mapping and a technical domain of the textual data using a pre-generated corpus. Furthermore, themethod comprises extracting using the one or more trained Al models, contextually intelligent information supplementing the textual data based on the type of mapping and the technical domain of the textual data using at least one of the pre-generated corpus, the textual data, and external data accessed from one or more external data sources associated with the computing system. Thereafter, the method comprises generating a second set of candidate phrases based on correlation of the extracted contextually intelligent information and the first set of candidate phrases. The second set of candidate phrases comprises phrases different from the first set of candidate phrases. Finally, the method comprises generating a report comprising one or more relevant technical documents conformant with the input data by mapping at least one of the first set of candidate phrases and the second set of candidate phrases with one or more technical documents of the technical domain based on the type of mapping and the extracted contextually intelligent information.

[0026] In an embodiment, the proposed method receives inputs in multiple formats such as text, image, video, and audio. This avoids the manual task of extracting texts from image, video and audio and allows users to directly provide an input to perform mapping. For instance, if a product manual is given as input and the product manual comprises links to a video within the text, the present disclosure ensures reviewing and extracting relevant text from the video as well along with the text content in the product manual. Hence, the present disclosure provides a method wherein combined effect of all data formats can be utilized to generate technical keywords, which makes the mapping more robust, accurate and meaningful. In an embodiment, the corpus used in the proposed method is manually checked and validated to identify accurate technical keywords from the list of frequently occurring words during the initial phases of generating the corpus, to ensure that the corpus generates valid technical words. In an embodiment, the proposed method also provides component to resource mapping which helps in mapping technical documents related to components disclosed in the input provided by the user. For instance, in the product manual, if the component discussed is about certain form of chip used in the laptop, the present disclosure enables identifying words related to such components such as microprocessor chip numbers, number of processor cores and the like, which are characteristics of the chip, and hence detects documents related to such technical keywords related to components to perform component to component mapping. This is possible due to the robust corpus of components which is built at the backend using at least one of an Al models, manual input, and public resources. In some embodiments, the component to resource mapping helps with activities such as infringement searches, invalidation searches,plagiarism searches and the like, where specifics of components used in a certain product are important. The existing techniques fail to disclose any method that can perform component to component mapping as these existing techniques fail to even recognize components such as microprocessor chips etc., as technical words based on their numbers and the like. In an embodiment, the proposed method extracts contextually intelligent information supplementing the textual data based on the type of mapping and the technical domain of the textual data which helps in generating the second set of candidate phrases. The second set of candidate phrases are relevant phrases or synonyms of the phrases present in textual data. The second set of candidate phrases also provide context of the textual data which helps in better mapping of the textual data with one or more relevant documents. Further, the contextually intelligent information is used to map textual data with one or more relevant documents which helps in enhancing the mapping which is performed which helps in providing efficient results.

[0027] In the following detailed description of the embodiments of the disclosure, reference is made to the accompanying drawings that form a part hereof, and in which are shown by way of illustration specific embodiments in which the disclosure may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the disclosure, and it is to be understood that other embodiments may be utilized and that changes may be made without departing from the scope of the present disclosure. The following description is, therefore, not to be taken in a limiting sense.

[0028] FIG. 1A shows an exemplary environment for mapping technical documents for a user query, in accordance with some embodiments of the present disclosure.

[0029] Exemplary architecture 100 comprises a computing system 101, a user device 103 a corpus 107, external data source 109i to external data source 109N (also referred as one or more external data sources 109N) and external databases 111. In an embodiment, a computing system 101 may be configured to identify technical documents for a user query. As an example, the computing system 101 may include, without limiting to, a desktop computer, a laptop, a mobile device and the like. As an example, the technical document may include, a patent document, a scientific paper, a remapping paper and the like that comprises technical information related to a product or a process. The computing system 101 may be associated with a user device 103. As an example, the user device 103 may include, without limiting to, a desktop computer, a laptop, a mobile device, and the like. In some embodiments, the computing system 101 may be a centralized system that receives input from a plurality of user devices (not shown in figure).The user device 103 and the computing system 101 may be connected using a communication network 105. As an example, the communication network 105 may be wireless communication network and / or wired communication network. In an embodiment, the computing system 101 may also be associated with the one or more external data sources 109N and external databases 111. The one or more external data sources 109N may be used to extract contextually intelligent information which will be discussed in detail in the further sections of the present disclosure. As an example, the one or more external data sources 109N may include, without limitation, web page, a scholarly article, a patent literature and any other publicly available resource. The external databases 111 may include one or more technical documents which may be used by the computing system 101 to map input data with the one or more technical documents. As an example, the one or more technical documents may include, without limitation, web page, a scholarly article, a patent literature, and any other publicly available resource. In some other embodiments, user device 103 may be locally connected with the computing system 101.

[0030] In an embodiment, the computing system 101 may be associated with the corpus 107 which may be used to identify technical documents for the user query. In an embodiment, the corpus 107 comprises plurality of technical phrases of one or more technical domains. The corpus 107 may be used to train one or more Al models which may be used to map technical documents for a user query. The method for generating corps is discussed in FIG. IF.

[0031] In an embodiment, the computing system 101 may be configured to receive a mapping input comprising input data in at least one data form of a plurality of data forms and a type of mapping to be performed. The plurality of forms of data may include, without limitation, a text form, an image form, a video form, and an audio form. Further, the type of mapping may include, without limitation, product to resource mapping, a resource to product mapping, a component to resource mapping and a resource to the component mapping, and a combinationbased mapping. At least one of the products, the resource and the component belong to at least one of the plurality of data forms. In an embodiment, in the product to resource mapping, the user may provide the product details and the computing system 101 may identify one or more resources which are related to the product details provided by the user. The one or more resources belong to at least one of the plurality of data forms. As an example, the resource may be web page, a scholarly article, a patent literature, and any other publicly available resource of any of the plurality of data forms. As an example, the product may be any apparatus / device for instance, being patented, marketed, sold, and the like. In an embodiment, in the resource toproduct mapping, the user may provide the resource such as a patent literature and the computing system 101 may identify one or more products which are related to the resource details provided by the user. As an example, if the user provides a patent specification related to a product “smart TV,” the computing system 101 may identify one or more products related to the smart TV that have same or similar features as described in the patent specification. In an embodiment, in the component to resource mapping, the user may provide a product manual comprising specific component details used in the product. As an example, the component may be any component within apparatus / device or a component by itself, for instance, being patented, marketed, sold, and the like. For instance, if the product is a smart TV, the component details may include for instance, type of processor, type of screen, type and specification of speaker and the like. The computing system 101 may identify one or more resources which are related to the same or similar product with the component details as specified in the product manual. The one or more resources belong to at least one of the plurality of data forms. In some other embodiments, the computing system 101 may identify one or more resources that are related specifically to the component itself such as a type of processor mentioned. In an embodiment, in the resource to component mapping, the user may provide the resource such as a patent literature and the computing system 101 may identify one or more components which are related to the resource details provided by the user. As an example, if the user provided a patent literature related to a SSD, the computing system 101 may identify one or more relevant technical documents related to one or more components in the SSD. In some embodiments, the user may select combination-based mapping which may include combination of the type of mapping.

[0032] In an embodiment, the user may be provided with options to provide input data and select the type of mapping to be performed on a Human Machine Interface (HMI) and / or a user interface on the user device. For instance, the user interface may be a web page, an application, and the like. The user may provide the input data in at least one data form of the plurality of data forms. For instance, the user may upload the input data and / or provide Uniform Resource Locator (URL) of the input data. Further, the user may select the type of mapping to be performed based on the plurality of options. As an example, the options for type of mapping may be provided in a drop down. Referring FIG. IB, the user may provide the input data in the text form, the image form, the video form, and the audio form. In some embodiments, the user may provide the input data in a data form which may further include data in the one or more other data forms. For instance, as shown in FIG. 1C, the input data provided by the useris a product description for a “Smart robot for cleaning home”. As shown in the figure, the input data is in a text form. However, the document comprises data in one or more data forms. As an example, the text form of data comprises an image of the smart cleaning robot, a video enclosure that provides video related to the smart cleaning robot, and also a hyperlink that redirects to a webpage comprising additional information related to the smart cleaning robot.

[0033] In an embodiment, upon receiving the mapping input, the computing system 101 may extract textual data from the input data received in the at least one data form of the plurality of data forms using one or more text extraction techniques. In an embodiment, the computing system 101 may detect each data form among the plurality of data forms present in the input data. As an example, the one or more text extraction techniques may include, without limitation, Optical Character Recognition (OCR), Computer Vision (CV) and Speech to Text. The OCR technique may be used to extract textual data from the input data in text form and image form, the CV technique may be used to extract textual data from the input data in image form and video form, and Speech to Text technique may be used to extract textual data from the input data in audio form. The extraction techniques are not limited to above example, as any other well-known techniques may also be used to extract the textual data. Upon detecting the data form, the computing system 101 may detect in the text form of input data, presence of reference links. The reference links may redirect towards a resource comprising additional input data of at least one of the plurality of data forms. Referring to FIG. 1C, the product description comprises a reference link which provides data related to features and technical details related to the product. Upon detecting the presence of reference links, the computing system 101 may extract using one or more text extracting techniques, the textual data from each of the plurality of detected data forms present in the input data, and the textual data from the additional input data obtained through the reference links, when the reference links are detected in the input data. As an example, the reference links may be detected by using predefined regular expressions. In an embodiment, the computing system 101 may perform data cleaning upon extracting the textual data from the input data.

[0034] During data cleaning, the computing system 101 may remove irrelevant characters such as punctuation marks, hashtags, and the like. Further, the computing system 101 may correct spelling mistakes and acronyms if any in the textual data. Thereafter, the computing system 101 may replace the emoji icon with the corresponding texting related to the emoji. In an embodiment, upon performing the data cleaning, the computing system 101 may perform datapre-processing. During the data pre-processing, the computing system 101 may perform, without limitation, at least one of, stemming, lemmatization, tokenization, and POS tagging. In some embodiments, the computing system 101 may select nouns and verbs for further processing. However, it is limited to the nouns and the verbs.

[0035] In an embodiment, upon extracting the textual data, the computing system 101 may determine using one or more trained Artificial Intelligence (Al) models, a first set of candidate phrases from the textual data based on the type of mapping and a technical domain of the textual data using a pre-generated corpus. The process associated with generating the corpus 107 is discussed in FIG. ID. As an example, the technical domain may be, telecommunication, metallurgy, pharma, biotechnology, cardiography, automobile, construction, architecture, Al, and Internet of Things (loT). The first set of candidate phrases technical key phrases related to the technical domain of the textual data. In an embodiment, the technical key phrases may include one or more words. In an embodiment, the computing system 101 may select the pregenerated corpus and trained Al model from the one or more trained Al models based on the type of mapping and the technical domain of the textual data. In an embodiment, during the training phase of the one or more Al models, the Al models may be trained based on the pregenerated corpus of different technical domains. In an embodiment, the one or more Al models may include the Al model for each type of mapping. As an example, if the user has selected product to resource mapping, the computing system 101 may select the Al model which may support the product to resource mapping. Further, the Al model is also trained with the pregenerated corpus which supports the product to resource mapping. Similarly, each Al model in the one or more Al models may be trained using the corresponding pre-generated corpus for each type of mapping. In some embodiments, the one or more trained Al models may include the Al model which may support all types of mapping. Upon selecting the pre-generated corpus and the trained Al model, the computing system 101 may determine the first set of candidate phrases from the textual data based on the technical domain of the textual data. In an embodiment, the computing system 101 may further compare the textual data with one or more technical key phrases in the selected pre-generated corpus to determine candidate phrases not determined by the selected trained Al model. This is performed to determine if the selected Al model has missed to determine any candidate phrases which may be relevant while performing mapping. If such candidate phrases are determined, the computing system 101 may append the additional candidate phrases to the first set of candidate phrases. In some other embodiments, determining frequently occurring words in the extracted text may be one of the methods ofdetermining the first set of candidate phrases. In some embodiments, the computing system 101 may compute an inverse of the frequency value for each word in the textual data. As an example, the computing system 101 may use Term Frequency - Inverse Document Frequency (TF-IDF) to identify the frequently occurring words. In some embodiments, apart from the frequently occurring words, the computing system 101 may identify acronyms and perform entity recognition using Named Entity Recognition (NER) technique. The NER technique is a Natural Language Processing (NLP) method that extracts information from text, detects and categorizes important information in text known as named entities. The computing system 101 may flag such NER and acronyms as relevant words from documents as well. In an embodiment, the computing system 101 may include these words in the first set of candidate phrases to perform further processing.

[0036] In an embodiment, upon determining first set of candidate phrases, the computing system 101 may extract using the one or more trained Al models, contextually intelligent information supplementing the textual data based on the type of mapping and the technical domain of the textual data using at least one of the pre-generated corpus, the textual data, and external data accessed from one or more external data sources 109N associated with the computing system 101. As an example, the one or more external data sources 109N may include, without limitation, scholarly articles, online databases, patent databases and any publicly available online resource. In an embodiment, the contextually intelligent information may be additional information related to the extracted textual data, that helps in understanding context of the textual data such as background, relevant content, related entities, and the like. In some embodiments, the contextually intelligent information may also be summary of the textual data, information associating the textual data with the author / applicant of the textual data, mapping textual data with images in the mapping input and any other relevant information that supplements the textual data contextually. For instance, if textual data relates to a mobile phone, the summary of the features of the mobile phone, company which owns the phone, the release date of the mobile phone and different variants of the mobile phone may be the contextually intelligent information which may be extracted. Consider an exemplary scenario in which the textual data relates to a cleaning device for a display. The first set of candidate phrases determined from the textual data may be the technical phrases present in the textual data. Further, the contextually intelligent information may be extracted which may provide a summary of the product description, the manufacturer of the cleaning device, the patent owner of the cleaning device and the features of the cleaning device. Also, the context of the cleaningdevice is a device used for cleaning the display and the contextually intelligent information extracted may supplement the textual data for a cleaning device for display. In another example, if the textual data relates to a cleaning device for floor of a home. The first set of candidate phrases determined from the textual data may be the technical phrases present in the textual data. Further, the contextually intelligent information may be extracted which may provide a summary of the product description, the manufacturer of the cleaning device, the patent owner of the cleaning device and the features of the cleaning device. Also, the context of the cleaning device is a device used in a home and the contextually intelligent information extracted may supplement the textual data for a cleaning device for home. In both the examples, though the products were related to cleaning device, the contextually intelligent information extracted from the textual data may be different for both the products. The contextually intelligent information helps to differentiate the products when there is same usage of terms. In an embodiment, the contextually intelligent information may be used to generate the second set of candidate phrases which is discussed below.

[0037] In an embodiment, upon extracting the contextually intelligent information, the computing system 101 may generate a second set of candidate phrases based on correlation of the extracted contextually intelligent information and the first set of candidate phrases. The second set of candidate phrases may include phrases different from the first set of candidate phrases. As discussed above, the computing system 101 may use the contextually intelligent information to generate the second set of candidate phrases. As an example, if the textual data relates to a cleaning device for a display. The first set of candidate phrases determined may be terms extracted from the textual data. Further, the contextually intelligent information may be extracted which provide details such as manufacturer of the product, where the product is used, target user of the product. For this example, the manufacturer may be company A, the product is used on display devices, the product may be used by a user having a display device. Here the context of the cleaning device is a device used for cleaning the display and the second set of candidate phrases generated based on correlation of the extracted contextually intelligent information and the first set of candidate phrases may be “screen cleaner,” “display cleaner,” “dust remover for screen” and the like. The contextually intelligent information helped in understanding the context of the textual data which helped in generating second set of candidate phrases which are related to the context of the textual data. Here the second set of candidate phrases may include phrases different from the first set of candidate phrases as the second set of candidate phrases are generated based on correlation of the extracted contextually intelligentinformation and the first set of candidate phrases. The candidate phrases in the second set of candidate phrases may not be directly available in the textual data and the first set of candidate phrases. In some embodiments, the computing system 101 may use one or more linguistic analysis methods to identify relevant words in the textual data. The one or more linguistic analysis methods may include, without limitation, at least one of, morphological analysis, semantic processing, and syntactic level examining. In an embodiment, the morphological analysis helps in identifying scientific study, identification, analysis, and description of the structure or form of words in a language and the relationship of the words to one another. In an embodiment, the semantic processing helps in determining possible meanings by activating interactions between meanings at the word level in the sentence. This level of processing may include the semantic uniqueness of words with multiple meanings. In an embodiment, the syntactic level examining emphasizes examining the words in a sentence to reveal the grammatical structure of the sentence. At this level, grammar and parsers are required. The result of this level of processing is the representation of the sentence, revealing the structural dependency relationships between the words. In an embodiment, the computing system 101 may include the phrases comprising words identified post performing the one or more linguistic analysis methods in the second set of candidate phrases to perform further processing.

[0038] In an embodiment, upon generating the second set of candidate phrases, the computing system 101 may generate a report comprising one or more relevant technical documents conformant with the input data by mapping at least one of the first set of candidate phrases and the second set of candidate phrases with one or more technical documents of the technical domain based on the type of mapping and the extracted contextually intelligent information. As an example, the one or more technical documents may include, without limitation, web page, a scholarly article, a patent literature, and any other publicly available resource. In an embodiment, the one or more relevant technical documents conformant with the input data states that the one or more relevant technical documents may be exactly similar with the input data. In some embodiments, the one or more relevant technical documents may include at least one of the first set of candidate phrases and the second set of candidate phrases, and the technical domain of the one or more relevant technical documents is similar to the technical domain of the input data. In an embodiment, the computing system 101 may generate result list corresponding to the identified one or more technical documents. The result list may include first list and second list. The first list may include the one or more technical documents identified corresponding to each candidate phrase in the at least one of the first set of candidatephrases and the second set of candidate phrases. The second result list may include, the one or more technical documents identified corresponding to more than one candidate phrase in the at least one of the first set of candidate phrases and the second set of candidate phrases. The one or more technical documents are arranged in an order of priority based on number of the second set of candidate phrases identified in the one or more technical documents. The one or more technical documents in the first result list and the second result list are assigned with a mapping score which indicates the closeness of the technical documents and the candidate phrases. In an embodiment, any known scoring techniques may be used to generate the mapping score . As shown in FIG. ID, the first result list is displayed to the user which indicates one or more technical documents identified corresponding to each candidate phrase in the at least one of the first set of candidate phrases. As shown in FIG. IE, the second result list is displayed to the user which indicates one or more technical documents identified corresponding to more than one candidate phrase in the at least one of the first set of candidate phrases and the second set of candidate phrases. In an embodiment, the user may provide feedback which indicates if the identified one or more technical documents are relevant to the first set of candidate phrases and the second set of candidate phrases. The user feedback may be used to train and enhance the one or more Al models.

[0039] FIG. IF shows an exemplary environment for generating a corpus 107, in accordance with some embodiments of the present disclosure.

[0040] In an embodiment, the computing system 101 (not shown in figure) may be configured to receive technical data related to a technical domain and corresponding ontology data in at least one data form among a plurality of data forms, from one or more data sources. The plurality of forms of data may include, without limitation, a text form, an image form, a video form, and an audio form. As an example, the one or more data sources may be, without limitation, web page, a scholarly article, a patent literature and any other publicly available resource. In an embodiment, the current corpus 107 may also be one of the data sources for generating / updating the corpus 107.

[0041] In an embodiment, upon receiving the technical data, the computing system 101 may extract textual data from the technical data and the corresponding ontology data using one or more text extraction techniques. As an example, the one or more text extraction techniques may include, without limitation, Optical Character Recognition (OCR), Computer V ision (CV) and Speech to Text. The OCR technique may be used to extract textual data from the input datain text form, the CV technique may be used to extract textual data from the input data in image form and video form, and Speech to Text technique may be used to extract textual data from the input data in audio form. The extractions techniques are not limited to above example, any other well-known techniques may also be used to extract the textual data from the technical data.

[0042] In an embodiment, upon extracting the textual data, the computing system 101 may determine a first set of technical corpus key phrases related to the technical domain from the extracted textual data using a trained Artificial Intelligence (Al) model. In an embodiment, the key phrases may include keywords and combination of keywords and key phrases. In an embodiment, the Al model may be initially trained using training data comprising a plurality of technical key phrases and ontology data related to the corresponding technical domain. In some embodiments, the training data may be provided manually during training of the Al model for first instance.

[0043] In an embodiment, upon determining first set of technical corpus key phrases, the computing system 101 may determine a second set of technical corpus key phrases related to the technical domain from the extracted textual data based on a user input. The second set of technical corpus key phrases are different from the first set of technical corpus key phrases. The user input may include, without limitation, manually annotating labels to the extracted textual data, domain based related phrases and dictionary based related phrases for each of the first set of technical corpus key phrases and second set of technical corpus key phrases. The domain based related phrases may be synonym or similar phrases in the technical domain of the technical data. As an example, the processor may also be known as a processing unit, a central processor in the computer science domain. The dictionary based related phrases may be synonym or similar phrases of the technical data. As an example, the processor may also be known as a Central Processing Unit (CPU). The user may manually identify the technical corpus key phrases which were not identified by the one or more trained Al model. As an example, the domain based related phrases and dictionary based related phrases may be synonymous related to the first set of technical corpus key phrases.

[0044] In an embodiment, upon determining second set of technical corpus key phrases, the computing system 101 may generate a domain specific knowledge graph indicating a relationship between the first set of technical corpus key phrases and the second set of technical corpus key phrases related to the technical domain. The domain specific knowledge graphprovides relationship between the first set of technical corpus key phrases and the second set of technical corpus key phrases which helps in understanding the relation between each key phrase in first set of technical corpus key phrases and the second set of technical corpus key phrases with the technical domain.

[0045] In an embodiment, upon generating the domain specific knowledge graph, the computing system 101 may generate a corpus 107 comprising technical corpus key phrases of the technical domain using the first set of technical corpus key phrases, the second set of technical corpus key phrases and the corresponding domain specific knowledge graph. In an embodiment, the corpus 107 may be used to train the one or more Al models. The corpus 107 may be updated periodically and the Al model may be trained each time the corpus 107 is updated. In an embodiment, similar steps may be used to generate corpus 107 related to each type of mapping and the technical domain. In some embodiments, the similar steps may be used to update the corpus 107 periodically.

[0046] FIG. 2 shows a detailed block diagram of the proposed computing system 101, in accordance with some embodiments of the present disclosure.

[0047] In some implementations, the computing system 101 may include an I / O interface 201, a processor 203 and a memory 205. In an embodiment, the memory 205 may be communicatively coupled to the processor 203. The processor 203 may be configured to perform one or more functions of a computing system 101 for mapping technical documents for a user query, using the data 207 and the one or more modules 209 of the computing system 101. In an embodiment, the memory 205 may store the data 207.

[0048] In an embodiment, the data 207 stored in the memory 205 may include, without limitation, mapping data 211 and corpus data 213. In an embodiment, the mapping data 211 may include, without limitation, input data 215, textual data 217, candidate phrases data 219, contextually intelligent information data 221 and other data 223. In an embodiment, the corpus data 213 may include, without limitation, technical data 225, textual data of corpus 227, technical corpus key phrases data 229 and other data 231. In some implementations, the data 207 may be stored within the memory 205 in the form of various data structures. Additionally, the data 207 may be organized using data models, such as relational or hierarchical data models. The other data 223 and 231 may include various temporary data and files generated by the one or more modules 209.

[0049] The mapping data 211 is discussed stored in the computing system 101 for mapping technical documents for a user query is discussed below.

[0050] In an embodiment, the input data 215 may include mapping input in at least one data form of a plurality of data forms provided by a user. The plurality of forms of data may include, without limitation, a text form, an image form, a video form, and an audio form. In some embodiments, the input data 215 may also include the type of mapping selected by the user. The type of mapping may include, without limitation, product to resource mapping, a resource to product mapping, a component to resource mapping and a resource to the component mapping, and a combination-based mapping. At least one of the products, the resource and the component belong to at least one of the plurality of data forms. In some embodiments, the input data 215 may be received via an Application Program Interface (API) call. As shown in FIG.IB, the user may provide the input data 215 in the text form, the image form, the video form, and the audio form. In some embodiments, the user may provide the input data 215 in one data form which may include data in the one or more data forms. For instance, as shown in FIG.IC, the input data 215 provided by the user is a product description for a “Smart robot for cleaning home”. As shown in the figure, the input data 215 is in a text form. However, the document comprises data in one or more data forms. As an example, the data is in image form, video form and hyperlink form.

[0051] In an embodiment, the textual data 217 may include the data extracted from the input data 215 using one or more text extraction techniques. As an example, the one or more text extraction techniques may include, without limitation, Optical Character Recognition (OCR), Computer Vision (CV) and Speech to Text. The OCR technique may be used to extract textual data 217 from the input data 215 in text form, the CV technique may be used to extract textual data 217 from the input data 215 in image form and video form, and Speech to Text technique may be used to extract textual data 217 from the input data 215 in audio form. The extractions techniques are not limited to above example, any other well-known techniques may also be used to extract the textual data 217. The textual data 217 may be in the text form and the textual data 217 may be further used to determine a first set of candidate phrases and second set of candidate phrases.

[0052] In an embodiment, the candidate phrases data 219 may include the first set of candidate phrases and a second set of candidate phrases. In an embodiment, the first set of candidate phrases may be determined using one or more trained Artificial Intelligence (Al) models, fromthe textual data 217 based on the type of mapping and a technical domain of the textual data 217 using a pre-generated corpus. The first set of candidate phrases may be directly extracted from the textual data 217 which are technical phrases related to the technical domain of the mapping input. In an embodiment, the second set of candidate phrases may be generated based on correlation of the extracted contextually intelligent information and the first set of candidate phrases. The second set of candidate phrases may be phrases related to the first set of candidate phrases which may be generated using the contextually intelligent information. As an example, the second set of candidate phrases may be synonyms of the first set of candidate phrases which may not be directly determined from the textual data 217. The second set of candidate phrases may include phrases different from the first set of candidate phrases. The candidate phrases data 219 may be further used to generate a report comprising one or more relevant technical documents conformant with the input data 215.

[0053] In an embodiment, the contextually intelligent information data 221 may be an information supplementing the textual data 217 which may be extracted based on the type of mapping and the technical domain of the textual data 217 using at least one of the pre-generated corpus, the textual data 217, and external data accessed from one or more external data sources 109N associated with the computing system 101. In an embodiment, contextually intelligent information data 221 may be used to generate the second set of candidate phrases based on correlation of the extracted contextually intelligent information and the first set of candidate phrases. In some embodiments, the contextually intelligent information data 221 may also be summary of the textual data 217, associating the textual data 217 with the author / applicant of the textual data 217 and mapping textual data 217 with images in the mapping input. For instance, if textual data 217 relates to a mobile phone, the summary of the features of the mobile phone, company which owns the phone, the release date of the mobile phone and different variants of the mobile phone may be the contextually intelligent information data 221 which may be extracted.

[0054] The corpus data 213 stored in the computing system 101 for generating corpus 107 is discussed below.

[0055] In an embodiment, the technical data 225 may include data related to a technical domain and corresponding ontology data in at least one data form among a plurality of data forms, from one or more data sources. The plurality of forms of data may include, without limitation, a text form, an image form, a video form, and an audio form. As an example, the one or more externaldata sources 109N may include, without limitation, scholarly articles, online databases, patent databases and any publicly available online resource.

[0056] In an embodiment, the textual data of the corpus 225 may include the data extracted from the technical data 225 using one or more text extraction techniques. As an example, the one or more text extraction techniques may include, without limitation, Optical Character Recognition (OCR), Computer Vision (CV) and Speech to Text. The OCR technique may be used to extract textual data 227 from the input data 215 in text form, the CV technique may be used to extract textual data 227 from the input data 215 in image form and video form, and Speech to Text technique may be used to extract textual data 227 from the input data 215 in audio form. The extractions techniques are not limited to above example, any other well-known techniques may also be used to extract the textual data 227. The textual data 227 may be in the text form and the textual data 227 may be further used to determine first set of technical corpus key phrases and second set of technical corpus key phrases.

[0057] In an embodiment, the technical corpus key phrases data 229 may include first set of technical corpus key phrases and second set of technical corpus key phrases. In an embodiment, the first set of technical corpus key phrases related to the technical domain are determined from the extracted textual data 227 using a trained Artificial Intelligence (Al) model. In an embodiment, the second set of technical corpus key phrases related to the technical domain may be determined from the extracted textual data 227 based on a user input. The user input may include, without limitation, manually annotating labels to the extracted textual data 227, domain based related phrases and dictionary based related phrases for each of the first set of technical corpus key phrases and second set of technical corpus key phrases. The domain based related phrases may be synonym or similar phrases in the technical domain of the technical data 225. As an example, the processor may also be known as a processing unit, a central processor in the computer science domain. The dictionary based related phrases may be synonyms or similar phrases of the technical data 225. As an example, the processor may also be known as a Central Processing Unit (CPU). In another example, vehicles driven without human interference may be referred in domain based related phrases as “autonomous vehicles,” “driverless vehicles,” “robotic vehicles” and the like, while in the dictionary based related phrases it may be only referred as “self-driving vehicle” or a “driverless vehicle.” The second set of technical corpus key phrases may be different from the first set of technical corpus key phrases. The first set of technical corpus key phrases and the second set of technical corpus keyphrases may be further used to generate a domain specific knowledge graph indicating a relationship between the first set of technical corpus key phrases and the second set of technical corpus key phrases related to the technical domain.

[0058] In an embodiment, the data 207 may be processed by the one or more modules 209 of the computing system 101. In some implementations, the one or more modules 209 may be communicatively coupled to the processor 203 for performing one or more functions of the computing system 101. In an implementation, the one or more modules 209 may include, without limiting to, a receiving module 233, an extracting module 235, a determining module 237, a generating module 239 and other modules 223.

[0059] As used herein, the term module may refer to an Application Specific Integrated Circuit (ASIC), an electronic circuit, a hardware processor (shared, dedicated, or group) and memory that execute one or more software or firmware programs, a combinational logic circuit, and / or other suitable components that provide the described functionality. In an implementation, each of the one or more modules 209 may be configured as stand-alone hardware computing unit. In an embodiment, the other modules 223 may be used to perform various miscellaneous functionalities on the computing system 101. It will be appreciated that such one or more modules 209 may be represented as a single module or a combination of different modules.

[0060] The modules are discussed below in view of method for mapping technical documents for a user query.

[0061] In an embodiment, the receiving module 233 may be configured to receiving a mapping input comprising input data 215 in at least one data form of a plurality of data forms and a type of mapping to be performed. The plurality of data forms may include, without limitation, a text form, an image form, a video form, and an audio form. The type of mapping may include, without limitation, product to resource mapping, a resource to product mapping, a component to resource mapping and a resource to the component mapping, and a combination-based search. At least one of the products, the resource and the component belong to at least one of the plurality of data forms.

[0062] In an embodiment, the extracting module 235 may be configured to extract textual data 217 from the input data 215 received in the at least one data form of the plurality of data forms using one or more text extraction techniques. In an embodiment, the extracting module 235 may detect each data form among the plurality of data forms present in the input data. Further,the extracting module 235 may detect in the text form of input data, presence of reference links, wherein the reference links redirect towards a resource comprising additional input data of at least one of the plurality of data forms. Upon detecting the presence of reference links, the extracting module 235 may extract using one or more text extracting techniques, the textual data from each of the plurality of detected data forms present in the input data, and the textual data from the additional input data obtained through the reference links, when the reference links are detected in the input data.

[0063] In an embodiment, the determining module 237 may be configured to determine using one or more trained Artificial Intelligence (Al) models, a first set of candidate phrases from the textual data 217 based on the type of mapping and a technical domain of the textual data 217 using a pre-generated corpus. In an embodiment, the determining module 237 may select the pre-generated corpus and trained Al model from the one or more trained Al models based on the type of mapping and the technical domain of the textual data. The selected trained Al model is trained based on the selected pre-generated corpus during a training phase. Further, the determining module 237 may determine using the selected trained Al model, the first set of candidate phrases from the textual data based on the technical domain of the textual data. In some embodiments, the determining module 237 may compare the textual data with one or more technical key phrases in the selected pre-generated corpus to determine candidate phrases not determined by the selected trained Al model. Further, the determining module 237 may append the additional candidate phrases to the first set of candidate phrases.

[0064] In an embodiment, the extracting module 235 may be configured to extract using the one or more trained Al models, contextually intelligent information supplementing the textual data 217 based on the type of mapping and the technical domain of the textual data 217 using at least one of the pre-generated corpus, the textual data 217, and external data accessed from one or more external data sources 109N associated with the computing system 101.

[0065] In an embodiment, the generating module 239 may be configured to generate a second set of candidate phrases based on correlation of the extracted contextually intelligent information and the first set of candidate phrases. The second set of candidate phrases comprises phrases different from the first set of candidate phrases.

[0066] In an embodiment, the generating module 239 may be configured to generate a report comprising one or more relevant technical documents conformant with the input data 215 bymapping at least one of the first set of candidate phrases and the second set of candidate phrases with one or more technical documents of the technical domain based on the type of mapping and the extracted contextually intelligent information. In an embodiment, the report may be displayed to the user on the display device associated with the user device. In an embodiment, the generating module 239 may generate result list corresponding to the identified one or more technical documents. The result list comprises at least one of a first result list comprising the one or more technical documents identified corresponding to each candidate phrase in the at least one of the first set of candidate phrases and the second set of candidate phrases and a second result list comprising the one or more technical documents identified corresponding to more than one candidate phrase in the at least one of the first set of candidate phrases and the second set of candidate phrases. The one or more technical documents are arranged in an order of priority based on number of the second set of candidate phrases identified in the one or more technical documents. The one or more technical documents in the first result list and the second result list are assigned with a mapping score which indicates the closeness of the technical documents and the candidate phrases.

[0067] The modules are discussed below in view of method for generating a corpus 107.

[0068] In an embodiment, the receiving module 233 may be configured to receive a mapping input comprising input data 215 in at least one data form of a plurality of data forms and a type of mapping to be performed.

[0069] In an embodiment, the extracting module 235 may be configured to extract textual data 227 from the technical data 225 and the corresponding ontology data using one or more text extraction techniques.

[0070] In an embodiment, the determining module 237 may be configured to determine a first set of technical corpus key phrases related to the technical domain from the extracted textual data 227 using a trained Artificial Intelligence (Al) model.

[0071] In an embodiment, the determining module 237 may be configured to determine using a second set of technical corpus key phrases related to the technical domain from the extracted textual data 227 based on a user input. The second set of technical corpus key phrases are different from the first set of technical corpus key phrases.

[0072] In an embodiment, the generating module 239 may be configured to generate a domain specific knowledge graph indicating a relationship between the first set of technical corpus key phrases and the second set of technical corpus key phrases related to the technical domain.

[0073] In an embodiment, the generating module 239 may be configured to generate a corpus 107 comprising technical corpus key phrases of the technical domain using the first set of technical corpus key phrases, the second set of technical corpus key phrases and the corresponding domain specific knowledge graph.

[0074] Consider an exemplary scenario in which a user provides a mapping input related to a product description of a smart home cleaning robot and selects product to resource mapping under type of mapping option. The textual data 217 extracted from the mapping data 211 is shown below:“Rethink the way you clean your home. Smart cleaning robot with vacuum feature can clean your floors whether you are at home or not. The smart robot comprises water tank as well to clean the floors post vacuum. The smart cleaning robot comprises a numerous sensors such as motion sensor, obstacle detection sensor and gyroscopic sensor. The smart cleaning robot covers twenty square feet each minute. An alarm is used to alert the user periodically to clean the home. The smart cleaning robot is processed by latest processor i 12.”

[0075] Upon extracting the textual data 217, the computing system 101 may determine the first set of candidate phrases and further extract contextually intelligent information supplementing the textual data 217. Thereafter, the second set of candidate phrases may be generated. The exemplary Table A below shows exemplary first set of candidate phrases, the contextually intelligent information supplementing the textual data 217 and the second set of candidate phrases:Table A

[0076] Finally, the computing system 101 may generate a report comprising one or more relevant technical documents conformant with the input data 215 by mapping at least one of the first set of candidate phrases and the second set of candidate phrases with one or more technical documents of the technical domain based on the type of mapping and the extracted contextually intelligent information. In an embodiment, the images of the input data may alsobe compared with the images in the one or more relevant documents. An exemplary table with list of one or more relevant documents is shown below:Table B

[0077] In another exemplary scenario when a user selects component to resource mapping for the same textual data 217, the computing system 101 may map the components and generate one or more relevant technical documents related to each component within the product. The one or more relevant documents generated in this type of mapping may identify documents related to each component within the product. In an embodiment, the images of the input data may also be compared with the images in the one or more relevant documents. An exemplary table with list of one or more relevant documents related to components within the product is shown below:Table C

[0078] FIG. 3 shows a flowchart illustrating a method of mapping technical documents for a user query, in accordance with some embodiments of the present disclosure.

[0079] As illustrated in FIG. 3, the method 300 may include one or more blocks illustrating a method of mapping technical documents for a user query. The method 300 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, and functions, which perform specific functions or implement specific abstract data types.

[0080] The order in which the method 300 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. Additionally, individual blocks may be deleted from the methods without departing from the scope of the subject matter described herein. Furthermore, the method can be implemented in any suitable hardware, software, firmware, or combination thereof.

[0081] At block 301, the method 300 includes receiving, by a processor 203 of a computing system 101, a mapping input comprising input data 215 in at least one data form of a plurality of data forms and a type of mapping to be performed. The plurality of forms of data may include, without limitation, a text form, an image form, a video form, and an audio form. The type of mapping may include, without limitation, product to resource mapping, a resource to product mapping, a component to resource mapping and a resource to the component mapping, and a combination-based mapping. At least one of the products, the resource and the component belong to at least one of the plurality of data forms.

[0082] At block 303, the method 300 includes extracting, by the processor 203, textual data 217 from the input data 215 received in the at least one data form of the plurality of data forms using one or more text extraction techniques. In an embodiment, to extract textual data 217, the processor may detect each data form among the plurality of data forms present in the input data 215. Further, the processor may detect in the text form of input data 215, presence of reference links, wherein the reference links redirect towards a resource comprising additional input data 215 of at least one of the plurality of data forms. Thereafter, the processor may extract using one or more text extracting techniques, the textual data 217 from each of the plurality of detected data forms present in the input data 215, and the textual data 217 from theadditional input data 215 obtained through the reference links, when the reference links are detected in the input data 215.

[0083] At block 305, the method 300 includes determining, by the processor 203, using one or more trained Artificial Intelligence (Al) models, a first set of candidate phrases from the textual data 217 based on the type of mapping and a technical domain of the textual data 217 using a pre-generated corpus. The one or more Al models are trained using a pre-generated corpus. In an embodiment, the processor may select the pre-generated corpus and trained Al model from the one or more trained Al models based on the type of mapping and the technical domain of the textual data 217. The selected trained Al model is trained based on the selected pregenerated corpus during a training phase. Further, the processor may determine using the selected trained Al model, the first set of candidate phrases from the textual data 217 based on the technical domain of the textual data 217. In an embodiment, the processor may further compare the textual data 217 with one or more technical key phrases in the selected pregenerated corpus to determine candidate phrases not determined by the selected trained Al model. Further, the processor may append the additional candidate phrases to the first set of candidate phrases.

[0084] At block 307, the method 300 includes extracting, by the processor 203, using the one or more trained Al models, contextually intelligent information supplementing the textual data 217 based on the type of mapping and the technical domain of the textual data 217 using at least one of the pre-generated corpus, the textual data 217, and external data accessed from one or more external data sources 109N associated with the computing system 101.

[0085] At block 309, the method 300 includes generating, by the processor 203, a second set of candidate phrases based on correlation of the extracted contextually intelligent information and the first set of candidate phrases. The second set of candidate phrases comprises phrases different from the first set of candidate phrases.

[0086] At block 311, the method 300 includes generating, by the processor 203, a report comprising one or more relevant technical documents conformant with the input data 215 by mapping at least one of the first set of candidate phrases and the second set of candidate phrases with one or more technical documents of the technical domain based on the type of mapping and the extracted contextually intelligent information. In an embodiment, the processor may generate result list corresponding to the identified one or more technical documents, whereinthe result list may include, at least one of a first result list comprising the one or more technical documents identified corresponding to each candidate phrase in the at least one of the first set of candidate phrases and the second set of candidate phrases and a second result list comprising the one or more technical documents identified corresponding to more than one candidate phrase in the at least one of the first set of candidate phrases and the second set of candidate phrases. The one or more technical documents are arranged in an order of priority based on number of the second set of candidate phrases identified in the one or more technical documents. The one or more technical documents in the first result list and the second result list are assigned with a mapping score which indicates the closeness of the technical documents and the candidate phrases.

[0087] FIG. 4 shows a flowchart illustrating a method of generating a corpus 107, in accordance with some embodiments of the present disclosure.

[0088] As illustrated in FIG. 4, the method 400 may include one or more blocks illustrating a method of generating a corpus 107. The method 400 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, and functions, which perform specific functions or implement specific abstract data types.

[0089] The order in which the method 400 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. Additionally, individual blocks may be deleted from the methods without departing from the scope of the subject matter described herein. Furthermore, the method can be implemented in any suitable hardware, software, firmware, or combination thereof.

[0090] At block 401, the method 400 includes receiving, by a processor 203 of a computing system 101, technical data 225 related to a technical domain and corresponding ontology data in at least one data form among a plurality of data forms, from one or more data sources.

[0091] At block 403, the method 400 includes extracting, by the processor 203, textual data 227 from the technical data 225 and the corresponding ontology data using one or more text extraction techniques.

[0092] At block 405, the method 400 includes determining, by the processor 203, a first set of technical corpus key phrases related to the technical domain from the extracted textual data using a trained Artificial Intelligence (Al) model. In an embodiment, the Al model may be initially trained using training data comprising a plurality of technical key phrases and ontology data related to the corresponding technical domain. The trained Al model may be an ensemble Al model.

[0093] At block 407, the method 400 includes determining, by the processor 203, a second set of technical corpus key phrases related to the technical domain from the extracted textual data based on a user input, wherein the second set of technical corpus key phrases are different from the first set of technical corpus key phrases. The user input may include, without limitation, manually annotating labels to the extracted textual data, domain based related phrases and dictionary based related phrases for each of the first set of technical corpus key phrases and second set of technical corpus key phrases.

[0094] At block 409, the method 400 includes generating, by the processor 203, a domain specific knowledge graph indicating a relationship between the first set of technical corpus key phrases and the second set of technical corpus key phrases related to the technical domain.

[0095] At block 411, the method 400 includes generating, by the processor 203, a corpus 107 comprising technical corpus key phrases of the technical domain using the first set of technical corpus key phrases, the second set of technical corpus key phrases and the corresponding domain specific knowledge graph.Computer System

[0096] FIG. 5 illustrates a block diagram of an exemplary computer system 500 for implementing embodiments consistent with the present disclosure. In an embodiment, the computer system 500 may be the computing system 101 that identifies technical documents for a user query and generates a corpus. The computer system 500 may include a central processing unit (“CPU” or “processor” or “memory controller”) 502. The processor 502 may comprise at least one data processor for executing program components for executing user- or systemgenerated business processes. A user may include a network manager, an application developer, a programmer, an organization, or any system / sub-system being operated parallelly to the computer system 500. The processor 502 may include specialized processing units suchas integrated system (bus) controllers, memory controllers / memory management computing systems, floating point units, graphics processing units, digital signal processing units, etc.

[0097] The processor 502 may be disposed in communication with one or more Input / Output (I / O) devices (511 and 512) via I / O interface 501. The I / O interface 501 may employ communication protocols / methods such as, without limitation, audio, analog, digital, stereo, IEEE®- 1394, serial bus, Universal Serial Bus (USB), infrared, PS / 2, BNC, coaxial, component, composite, Digital Visual Interface (DVI), high-definition multimedia interface (HDMI), Radio Frequency (RF) antennas, S-Video, Video Graphics Array (VGA), IEEE® 8O2.n / b / g / n / x, Bluetooth, cellular (e.g., Code-Division Multiple Access (CDMA), High-Speed Packet Access (HSPA+), Global System For Mobile Communications (GSM), Long-Term Evolution (LTE) or the like), etc. Using the I / O interface 501, the computer system 500 may communicate with one or more I / O devices 511 and 512.

[0098] In some embodiments, the processor 502 may be disposed in communication with a network 509 via a network interface 503. The network interface 503 may communicate with the network 509. The network interface 503 may employ connection protocols including, without limitation, direct connect, Ethernet (e.g., twisted pair 10 / 100 / 1000 Base T), Transmission Control Protocol / Intemet Protocol (TCP / IP), token ring, IEEE® 802.1 la / b / g / n / x, etc.

[0099] In an implementation, the preferred network 509 may be implemented as one of the several types of networks, such as intranet or Local Area Network (LAN) and such within the organization. The preferred network 509 may either be a dedicated network or a shared network, which represents an association of several types of networks that use a variety of protocols, for example, Hypertext Transfer Protocol (HTTP), Transmission Control Protocol / Intemet Protocol (TCP / IP), Wireless Application Protocol (WAP) etc., to communicate with each other. Further, the network 509 may include a variety of network devices, including routers, bridges, servers, computing devices, storage devices, etc. Using the network interface 503 and the network 509, the computer system 500 may communicate with user device 103 corpus 107, one or more external data sources 109N and external database 111.

[0100] In some embodiments, the processor 502 may be disposed in communication with a memory 505 (e.g., RAM 513, ROM 514, etc. as shown in FIG. 5) via a storage interface 504. The storage interface 504 may connect to memory 505 including, without limitation, memorydrives, removable disc drives, etc., employing connection protocols such as Serial Advanced Technology Attachment (SATA), Integrated Drive Electronics (IDE), IEEE- 1394, Universal Serial Bus (USB), fiber channel, Small Computer Systems Interface (SCSI), etc. The memory drives may further include a drum, magnetic disc drive, magneto-optical drive, optical drive, Redundant Array of Independent Discs (RAID), solid-state memory devices, solid-state drives, etc.

[0101] The memory 505 may store a collection of program or database components, including, without limitation, user / application interface 506, an operating system 507, a web browser 508, and the like. In some embodiments, computer system 500 may store user / application data 506, such as the data, variables, records, etc. as described in this invention. Such databases may be implemented as fault-tolerant, relational, scalable, secure databases such as Oracle® or Sybase®.

[0102] The operating system 507 may facilitate resource management and operation of the computer system 500. Examples of operating systems include, without limitation, APPLE® MACINTOSH® OS X®, UNIX®, UNIX-like system distributions (E G., BERKELEY SOFTWARE DISTRIBUTION® (BSD), FREEBSD®, NETBSD®, OPENBSD, etc ), LINUX® DISTRIBUTIONS (E G., RED HAT®, UBUNTU®, KUBUNTU®, etc ), IBM® OS / 2®, MICROSOFT® WINDOWS® (XP®, VISTA® / 7 / 8, 10 etc ), APPLE® IOS®, GOOGLETMANDROID ™, BLACKBERRY® OS, or the like.

[0103] The user interface 506 may facilitate display, execution, interaction, manipulation, or operation of program components through textual or graphical facilities. For example, the user interface 506 may provide computer interaction interface elements on a display system operatively connected to the computer system 500, such as cursors, icons, check boxes, menus, scrollers, windows, widgets, and the like. Further, Graphical User Interfaces (GUIs) may be employed, including, without limitation, APPLE® MACINTOSH® operating systems’ Aqua®, IBM® OS / 2®, MICROSOFT® WINDOWS® (e.g., Aero, Metro, etc.), web interface libraries (e g., ActiveX®, JAVA®, JAVASCRIPT®, AJAX, HTML, ADOBE® FLASH®, etc ), or the like.

[0104] The web browser 508 may be a hypertext viewing application. Secure web browsing may be provided using Secure Hypertext Transport Protocol (HTTPS), Secure Sockets Layer (SSL), Transport Layer Security (TLS), and the like. The web browsers 508 may utilizefacilities such as AJAX, DHTML, ADOBE® FLASH®, JAVASCRIPT®, JAVA®, Application Programming Interfaces (APIs), and the like. Further, the computer system 500 may implement a mail server stored program component. The mail server may utilize facilities such as ASP, ACTIVEX®, ANSI® C++ / C#, MICROSOFT®, NET, CGI SCRIPTS, JAVA®, JAVASCRIPT®, PERL®, PHP, PYTHON®, WEBOBJECTS®, etc. The mail server may utilize communication protocols such as Internet Message Access Protocol (IMAP), Messaging Application Programming Interface (MAPI), MICROSOFT® exchange, Post Office Protocol (POP), Simple Mail Transfer Protocol (SMTP), or the like. In some embodiments, the computer system 500 may implement a mail client stored program component. The mail client may be a mail viewing application, such as APPLE® MAIL, MICROSOFT® ENTOURAGE®, MICROSOFT® OUTLOOK®, MOZILLA® THUNDERBIRD®, and the like.

[0105] Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present invention. 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, Compact Disc (CD) ROMs, Digital Video Disc (DVDs), flash drives, disks, and any other known physical storage media.Advantages of the embodiments of the present disclosure are illustrated herein.

[0106] In an embodiment, the proposed method identifies technical phrases accurately which helps to find the most relevant technical documents.

[0107] In an embodiment, the proposed method receives inputs in multiple formats such as text, image, video, and audio. This avoids the manual task of extracting texts from image, video and audio and allows users to directly provide an input to perform mapping. For instance, if a product manual is given as input and the product manual comprises links to a video within the text, the present disclosure ensures reviewing and extracting relevant text from the video as well along with the text content in the product manual. Hence, the present disclosure providesa method wherein combined effect of all data formats can be utilized to generate technical keywords, which makes the mapping more robust, accurate and meaningful.

[0108] In an embodiment, the corpus used in the proposed method is manually checked and validated to identify accurate technical keywords from the list of frequently occurring words during the initial phases of generating the corpus, to ensure that the corpus generates valid technical words.

[0109] In an embodiment, the proposed method also provides component to component mapping which helps in mapping technical documents related to components disclosed in the input provided by the user. For instance, in the product manual, if the component discussed is about certain form of chip used in the laptop, the present disclosure enables identifying words related to such components such as microprocessor chip numbers, number of processor cores and the like, which are characteristics of the chip, and hence detects documents related to such technical keywords related to components to perform component to component mapping. This is possible due to the robust corpus of components which is built at the backend using at least one of an Al models, manual input, and public resources. In some embodiments, the component to re mapping helps with activities such as infringement searches, invalidation searches, plagiarism searches and the like, where specifics of components used in a certain product are important. The existing techniques fail to disclose any method that can perform component to component mapping as these existing techniques fail to even recognize components such as microprocessor chips etc., as technical words based on their numbers and the like.[HO] In light of the technical advancements provided by the disclosed method and system, the claimed steps, as discussed above, are not routine, conventional, or well-known aspects in the art, as the claimed steps provide the aforesaid solutions to the technical problems existing in the conventional technologies. Further, the claimed steps clearly bring an improvement in the functioning of the system itself, as the claimed steps provide a technical solution to a technical problem.[Hl] The terms "an embodiment", "embodiment", "embodiments", "the embodiment", "the embodiments", "one or more embodiments", "some embodiments", and "one embodiment" mean "one or more (but not all) embodiments of the invention(s)" unless expressly specified otherwise.

[0112] The terms "including", "comprising", “having” and variations thereof mean "including but not limited to", unless expressly specified otherwise.

[0113] The enumerated listing of items does not imply that any or all the items are mutually exclusive, unless expressly specified otherwise. The terms "a", "an" and "the" mean "one or more", unless expressly specified otherwise.

[0114] A description of an embodiment with several components in communication with each other does not imply that all such components are required. On the contrary, a variety of optional components are described to illustrate the wide variety of possible embodiments of the invention.

[0115] When a single device or article is described herein, it will be clear that more than one device / article (whether they cooperate) may be used in place of a single device / article. Similarly, where more than one device / article is described herein (whether they cooperate), it will be clear that a single device / article may be used in place of the more than one device / article or a different number of devices / articles may be used instead of the shown number of devices or programs. The functionality and / or features of a device may be alternatively embodied by one or more other devices which are not explicitly described as having such fimctionality / features. Thus, other embodiments of invention need not include the device itself.

[0116] Finally, the language used in the specification has been principally selected for readability and instructional purposes, and it may not have been selected to delineate or circumscribe the inventive subj ect matter. It is therefore intended that the scope of the invention be limited not by this detailed description, but rather by any claims that issue on an application based here on. Accordingly, the embodiments of the present invention are intended to be illustrative, but not limiting, of the scope of the invention, which is set forth in the following claims.

[0117] While various aspects and embodiments have been disclosed herein, other aspects and embodiments will be apparent to those skilled in the art. The various aspects and embodiments disclosed herein are for purposes of illustration and are not intended to be limiting, with the true scope and spirit being indicated by the following claims.

Claims

WE CLAIM:

1. A method of mapping technical documents for a user query, the method comprising: receiving, by a computing system (101), a mapping input comprising input data in at least one data form of a plurality of data forms and a type of mapping to be performed; extracting, by the computing system (101), textual data from the input data received in the at least one data form of the plurality of data forms using one or more text extraction techniques; determining, by the computing system (101) using one or more trained Artificial Intelligence (Al) models, a first set of candidate phrases from the textual data based on the type of mapping and a technical domain of the textual data using a pre-generated corpus; extracting, by the computing system (101) using the one or more trained Al models, contextually intelligent information supplementing the textual data based on the type of mapping and the technical domain of the textual data using at least one of the pre-generated corpus, the textual data, and external data accessed from one or more external data sources (109N) associated with the computing system (101); generating, by the computing system (101), a second set of candidate phrases based on correlation of the extracted contextually intelligent information and the first set of candidate phrases, wherein the second set of candidate phrases comprises phrases different from the first set of candidate phrases; and generating, by the computing system (101), a report comprising one or more relevant technical documents conformant with the input data by mapping at least one of the first set of candidate phrases and the second set of candidate phrases with one or more technical documents of the technical domain based on the type of mapping and the extracted contextually intelligent information.

2. The method as claimed in claim 1 , wherein the plurality of forms of data comprises a text form, an image form, a video form, and an audio form.

3. The method as claimed in claim 1, wherein extracting the textual data from the input data comprises:detecting each data form among the plurality of data forms present in the input data, wherein the plurality of data forms comprises a text form, an image form, a video form, and an audio form; detecting, in the text form of input data, presence of reference links, wherein the reference links redirect towards a resource comprising additional input data of at least one of the plurality of data forms; and extracting, using one or more text extracting techniques, the textual data from each of the plurality of detected data forms present in the input data, and the textual data from the additional input data obtained through the reference links, when the reference links are detected in the input data.

4. The method as claimed in claim 1, wherein determining the first set of candidate phrases further comprises: selecting the pre-generated corpus and trained Al model from the one or more trained Al models based on the type of mapping and the technical domain of the textual data, wherein the selected trained Al model is trained based on the selected pre-generated corpus during a training phase; and determining, using the selected trained Al model, the first set of candidate phrases from the textual data based on the technical domain of the textual data.

5. The method as claimed in claim 4, wherein determining the first set of candidate phrases further comprises: comparing the textual data with one or more technical key phrases in the selected pre-generated corpus to determine candidate phrases not determined by the selected trained Al model; and appending the additional candidate phrases to the first set of candidate phrases.

6. The method as claimed in claim 1 further comprises generating, by the computing system (101), result list corresponding to the identified one or more technical documents, wherein the result list comprises at least one of:a first result list comprising the one or more technical documents identified corresponding to each candidate phrase in the at least one of the first set of candidate phrases and the second set of candidate phrases; and a second result list comprising the one or more technical documents identified corresponding to more than one candidate phrase in the at least one of the first set of candidate phrases and the second set of candidate phrases, wherein the one or more technical documents are arranged in an order of priority based on number of the second set of candidate phrases identified in the one or more technical documents, wherein the one or more technical documents in the first result list and the second result list are assigned with a mapping score which indicates the closeness of the technical documents and the candidate phrases.

7. The method as claimed in claim 1, wherein the type of mapping comprises, product to resource mapping, a resource to product mapping, a component to resource mapping and a resource to the component mapping, and a combination based mapping, wherein at least one of the product, the resource and the component belong to at least one of the plurality of data forms.

8. The method as claimed in claim 1, wherein the one or more Al models are trained using a pre-generated corpus.

9. A method of generating a corpus, the method comprising: receiving, by a computing system (101), technical data related to a technical domain and corresponding ontology data in at least one data form among a plurality of data forms, from one or more data sources; extracting, by the computing system (101), textual data from the technical data and the corresponding ontology data using one or more text extraction techniques; determining, by the computing system (101), a first set of technical corpus key phrases related to the technical domain from the extracted textual data using a trained Artificial Intelligence (Al) model;determining, by the computing system (101), a second set of technical corpus key phrases related to the technical domain from the extracted textual data based on a user input, wherein the second set of technical corpus key phrases are different from the first set of technical corpus key phrases; generating, by the computing system (101), a domain specific knowledge graph indicating a relationship between the first set of technical corpus key phrases and the second set of technical corpus key phrases related to the technical domain; and generating, by the computing system (101), a corpus comprising technical corpus key phrases of the technical domain using the first set of technical corpus key phrases, the second set of technical corpus key phrases and the corresponding domain specific knowledge graph.

10. The method as claimed in claim 9, wherein the Al model is initially trained using training data comprising a plurality of technical key phrases and ontology data related to the corresponding technical domain.

11. The method as claimed in claim 9, wherein the user input comprises manually annotating labels to the extracted textual data, domain based related phrases and dictionary based related phrases for each of the first set of technical corpus key phrases and second set of technical corpus key phrases.

12. A computing system (101) for mapping technical documents for a user query, the computing system (101) comprising: a processor (203); and a memory (205), communicatively coupled to the processor (203), wherein the memory (205) stores processor (203) executable instructions, which, on execution, causes the processor (203) to: receive a mapping input comprising input data in at least one data form of a plurality of data forms and a type of mapping to be performed; extract textual data from the input data received in the at least one data form of the plurality of data forms using one or more text extraction techniques;determine, by the computing system (101) using one or more trained Artificial Intelligence (Al) models, a first set of candidate phrases from the textual data based on the type of mapping and a technical domain of the textual data using a pre-generated corpus; extract using the one or more trained Al models, contextually intelligent information supplementing the textual data based on the type of mapping and the technical domain of the textual data using at least one of the pre-generated corpus, the textual data, and external data accessed from one or more external data sources (109N) associated with the computing system (101); generate a second set of candidate phrases based on correlation of the extracted contextually intelligent information and the first set of candidate phrases, wherein the second set of candidate phrases comprises phrases different from the first set of candidate phrases; and generate a report comprising one or more relevant technical documents conformant with the input data by mapping at least one of the first set of candidate phrases and the second set of candidate phrases with one or more technical documents of the technical domain based on the type of mapping and the extracted contextually intelligent information.

13. The computing system (101) as claimed in claim 12, wherein extracting the textual data from the input data comprises: detecting each data form among the plurality of data forms present in the input data, wherein the plurality of data forms comprises a text form, an image form, a video form, and an audio form; detecting, in the text form of input data, presence of reference links, wherein the reference links redirect towards a resource comprising additional input data of at least one of the plurality of data forms; and extracting, using one or more text extracting techniques, the textual data from each of the plurality of detected data forms present in the input data, and the textual data from the additional input data obtained through the reference links, when the reference links are detected in the input data.

14. The computing system (101) as claimed in claim 12, wherein determining the first set of candidate phrases further comprises: selecting the pre-generated corpus and trained Al model from the one or more trained Al models based on the type of mapping and the technical domain of the textual data, wherein the selected trained Al model is trained based on the selected pre-generated corpus during a training phase; and determining, using the selected trained Al model, the first set of candidate phrases from the textual data based on the technical domain of the textual data.

15. The computing system (101) as claimed in claim 13, wherein to determine the first set of candidate phrases, the processor (203) is further configured to: compare the textual data with one or more technical key phrases in the selected pregenerated corpus to determine candidate phrases not determined by the selected trained Al model; and append the additional candidate phrases to the first set of candidate phrases.

16. The computing system (101) as claimed in claim 12 further comprises generating, by the computing system (101), result list corresponding to the identified one or more technical documents, wherein the result list comprises at least one of: a first result list comprising the one or more technical documents identified corresponding to each candidate phrase in the at least one of the first set of candidate phrases and the second set of candidate phrases; and a second result list comprising the one or more technical documents identified corresponding to more than one candidate phrase in the at least one of the first set of candidate phrases and the second set of candidate phrases, wherein the one or more technical documents are arranged in an order of priority based on number of the second set of candidate phrases identified in the one or more technical documents, wherein the one or more technical documents in the first result list and the second result list are assigned with a mapping score which indicates the closeness of the technical documents and the candidate phrases.

17. A computing system (101) of generating a corpus, the computing system (101) comprising: a processor (203); and a memory (205), communicatively coupled to the processor (203), wherein the memory (205) stores processor (203) executable instructions, which, on execution, causes the processor (203) to: receive technical data related to a technical domain and corresponding ontology data in at least one data form among a plurality of data forms, from one or more data sources; extract textual data from the technical data and the corresponding ontology data using one or more text extraction techniques; determine a first set of technical corpus key phrases related to the technical domain from the extracted textual data using a trained Artificial Intelligence (Al) model; determine a second set of technical corpus key phrases related to the technical domain from the extracted textual data based on a user input, wherein the second set of technical corpus key phrases are different from the first set of technical corpus key phrases; generate a domain specific knowledge graph indicating a relationship between the first set of technical corpus key phrases and the second set of technical corpus key phrases related to the technical domain; and generate a corpus comprising technical corpus key phrases of the technical domain using the first set of technical corpus key phrases, the second set of technical corpus key phrases and the corresponding domain specific knowledge graph.

18. The computing system (101) as claimed in claim 17, wherein the Al model is initially trained using training data comprising a plurality of technical key phrases and ontology data related to the corresponding technical domain.

19. The computing system (101) as claimed in claim 17, wherein the user input comprises manually annotating labels to the extracted textual data, domain based related phrases and dictionary based related phrases for each of the first set of technical corpus key phrases and second set of technical corpus key phrases.

Images