Claims:

1. A method of generating a real time contextual summary from analytical reports by machine learning techniques, the method comprising a processor implemented steps of:
capturing, by one or more hardware processors, a first set of information comprising a set of user inputs via one or more devices (301);
extracting, based upon the first set of information, a second set of information comprising user intents and entities corresponding to the first set of information via a cognitive engine, wherein the second set of information is extracted by implementing one or more natural language processing techniques (302);
generating, based upon the second set of information, a plurality of queries on one or more analytical reports to be generated corresponding to the first set of information (303);
performing, using the plurality of queries generated, a plurality of steps in parallel, wherein the plurality of steps comprise (304):
(i) generating the one or more analytical reports in a textual form corresponding to the first set of information (304(i)); and
(ii) determining a third set of information comprising statistical data corresponding to the one or more analytical reports generated (304(ii));
identifying, from one or more domain ontologies, a plurality of words associated with the third set of information via a machine learning module, wherein each of the plurality of words represents an analytical data corresponding to the real time contextual summary to be generated (305); and
generating, using each of the plurality of words identified, the real time contextual summary from the one or more analytical reports via the machine learning module, wherein the real time contextual summary is generated by implementing one or more Recurrent Neural Network (RNN) techniques (306).

2. The method of claim 1, wherein the step of generating the real time contextual summary comprises:
(i) obtaining, from the plurality of words, a fourth set of information comprising one or more domain-specific sentences corresponding to the real time contextual summary to be generated by implementing the one or more RNN techniques; and
(ii) generating, from the fourth set of information, the real time contextual summary comprising statistical and non-statistical interpretations from the one or more analytical reports generated.

3. The method of claim 2, wherein the step of obtaining the fourth set of information is preceded by identifying, based upon the third set of information, a sequence of representation of the fourth set of information to generate the real time contextual summary, and wherein the sequence of representation is identified by implementing the one or more RNN techniques.

4. The method of claim 3, wherein the step of identifying the sequence of representation comprises predicting an order of generation of each of the plurality of words by implementing the one or more RNN techniques.

5. A system (100) for generating a real time contextual summary from analytical reports by machine learning techniques, the system (100) comprising:
a memory (102) storing instructions;
one or more communication interfaces (106); and
one or more hardware processors (104) coupled to the memory (102) via the one or more communication interfaces (106), wherein the one or more hardware processors (104) are configured by the instructions to:
capture a first set of information comprising a set of user inputs via one or more devices (202);
extract, based upon the first set of information, a second set of information comprising user intents and entities corresponding to the first set of information via a cognitive engine; wherein the second set of information is extracted by implementing one or more natural language processing techniques;
generate, based upon the second set of information, a plurality of queries on one or more analytical reports to be generated corresponding to the first set of information;
perform, using the plurality of queries generated, a plurality of steps in parallel, wherein the plurality of steps comprise:
(i) generate the one or more analytical reports in a textual form corresponding to the first set of information; and
(ii) determine a third set of information comprising statistical data corresponding to the one or more analytical reports generated;
identify, from one or more domain ontologies (206), a plurality of words associated with the third set of information via a machine learning module (207), wherein each of the plurality of words represents an analytical data corresponding to the real time contextual summary to be generated; and
generate, using each of the plurality of words identified, the real time contextual summary from the one or more analytical reports via the machine learning module (207), wherein the real time contextual summary is generated by implementing one or more Recurrent Neural Network (RNN) techniques.

6. The system (100) of claim 5, wherein the one or more hardware processors (104) are configured to generate the real time contextual summary by:
(i) obtaining, from the plurality of words, a fourth set of information comprising one or more domain-specific sentences corresponding to the real time contextual summary to be generated by implementing the one or more RNN techniques; and
(ii) generating, from the fourth set of information, the real time contextual summary comprising statistical and non-statistical interpretations from the one or more analytical reports generated.

7. The system (100) of claim 6, wherein the one or more hardware processors (104) are configured to obtain the fourth set of information by identifying, based upon the third set of information, a sequence of representation of the fourth set of information to generate the real time contextual summary, and wherein the sequence of representation is identified by implementing the one or more RNN techniques.

8. The system (100) of claim 7, wherein the one or more hardware processors (104) are configured to identify the sequence of representation by predicting an order of generation of each of the plurality of words by implementing the one or more RNN techniques.
, Description:FORM 2

THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003

COMPLETE SPECIFICATION
(See Section 10 and Rule 13)

Title of invention:

SYSTEMS AND METHODS FOR GENERATING REAL TIME CONTEXTUAL SUMMARY FROM ANALYTICAL REPORTS BY MACHINE LEARNING

Applicant

Tata Consultancy Services Limited
A company Incorporated in India under the Companies Act, 1956
Having address:
Nirmal Building, 9th floor,
Nariman point, Mumbai 400021,
Maharashtra, India

The following specification particularly describes the invention and the manner in which it is to be performed.


TECHNICAL FIELD
[001] The disclosure herein generally relates to generating a real time contextual summary by machine learning, and, more particularly, to systems and methods for generating a real time contextual summary from one or more analytical reports by one or more machine learning techniques.

BACKGROUND
[002] Machine learning involves the development of tools that can extract knowledge from data sets once the representation for the data has been defined. The discovered knowledge may comprise of rules describing properties of the data, frequently occurring patterns, clustering of objects in the data set, etc. During machine learning, many steps may precede an actual model construction step, for example, preparing a training data, cleaning that data, applying the appropriate transformations to the data, eliminating misclassified data. Other steps follow such as evaluating the classification model and consolidating newly produced models with already existing models. Thus, to be effective, the actual model construction must be integrated into a machine leaning methodology that supports the entire process of transforming data to useful classification knowledge.
[003] Real time data analytics has become critical for organizations these days. Machine learning is a method of data analysis that automates analytical model building. It is a branch of artificial intelligence based on the idea that systems can learn from data, identify patterns and make decisions with minimal human intervention. Machine learning operates on the premise that computers learn from data; the computers and their models adapt independently as more data are fed to them. Machine learning is evolving as data mining becomes more critical to businesses who need to make sense of big data.

SUMMARY
[004] Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems. For example, in one embodiment, a method for generating a real time contextual summary from analytical reports by machine learning techniques is provided, the method comprising: capturing, by one or more hardware processors, a first set of information comprising a set of user inputs via one or more devices; extracting, based upon the first set of information, a second set of information comprising user intents and entities corresponding to the first set of information via a cognitive engine, wherein the second set of information is extracted by implementing one or more natural language processing techniques; generating, based upon the second set of information, a plurality of queries on one or more analytical reports to be generated corresponding to the first set of information; performing, using the plurality of queries generated, a plurality of steps in parallel, wherein the plurality of steps comprise: (i) generating the one or more analytical reports in a textual form corresponding to the first set of information; and determining a third set of information comprising statistical data corresponding to the one or more analytical reports generated; identifying, from one or more domain ontologies, a plurality of words associated with the third set of information via a machine learning module, wherein each of the plurality of words represents an analytical data corresponding to the real time contextual summary to be generated; generating, using each of the plurality of words identified, the real time contextual summary from the one or more analytical reports via the machine learning module, wherein the real time contextual summary is generated by implementing one or more Recurrent Neural Network (RNN) techniques; obtaining, from the plurality of words, a fourth set of information comprising one or more domain-specific sentences corresponding to the real time contextual summary to be generated by implementing the one or more RNN techniques; generating, from the fourth set of information, the real time contextual summary comprising statistical and non-statistical interpretations from the one or more analytical reports generated; identifying, based upon the third set of information, a sequence of representation of the fourth set of information to generate the real time contextual summary, wherein the sequence of representation is identified by implementing the one or more RNN techniques; and predicting an order of generation of each of the plurality of words by implementing the one or more RNN techniques for identifying the sequence of representation.
[005] In another aspect, there is provided a system for generating a real time contextual summary from analytical reports by machine learning techniques, the system comprising a memory storing instructions; one or more communication interfaces; and one or more hardware processors coupled to the memory via the one or more communication interfaces, wherein the one or more hardware processors are configured by the instructions to: capture a first set of information comprising a set of user inputs via one or more devices; extract, based upon the first set of information, a second set of information comprising user intents and entities corresponding to the first set of information via a cognitive engine; wherein the second set of information is extracted by implementing one or more natural language processing techniques; generate, based upon the second set of information, a plurality of queries on one or more analytical reports to be generated corresponding to the first set of information; perform, using the plurality of queries generated, a plurality of steps in parallel, wherein the plurality of steps comprise: (i) generate the one or more analytical reports in a textual form corresponding to the first set of information; and (ii) determine a third set of information comprising statistical data corresponding to the one or more analytical reports generated; identify, from one or more domain ontologies, a plurality of words associated with the third set of information via a machine learning module, wherein each of the plurality of words represents an analytical data corresponding to the real time contextual summary to be generated; generate, using each of the plurality of words identified, the real time contextual summary from the one or more analytical reports via the machine learning module, wherein the real time contextual summary is generated by implementing one or more Recurrent Neural Network (RNN) techniques; obtain, from the plurality of words, a fourth set of information comprising one or more domain-specific sentences corresponding to the real time contextual summary to be generated by implementing the one or more RNN techniques; generate, from the fourth set of information, the real time contextual summary comprising statistical and non-statistical interpretations from the one or more analytical reports generated; obtain the fourth set of information by identifying, based upon the third set of information, a sequence of representation of the fourth set of information to generate the real time contextual summary, and wherein the sequence of representation is identified by implementing the one or more RNN techniques; and identify the sequence of representation by predicting an order of generation of each of the plurality of words by implementing the one or more RNN techniques.
[006] In yet another aspect, there is provided one or more non-transitory machine readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors causes the one or more hardware processors to perform a method for generating a real time contextual summary from analytical reports by machine learning techniques, the method comprising: capturing a first set of information comprising a set of user inputs via one or more devices; extracting, based upon the first set of information, a second set of information comprising user intents and entities corresponding to the first set of information via a cognitive engine, wherein the second set of information is extracted by implementing one or more natural language processing techniques; generating, based upon the second set of information, a plurality of queries on one or more analytical reports to be generated corresponding to the first set of information; performing, using the plurality of queries generated, a plurality of steps in parallel, wherein the plurality of steps comprise: (i) generating the one or more analytical reports in a textual form corresponding to the first set of information; and determining a third set of information comprising statistical data corresponding to the one or more analytical reports generated; identifying, from one or more domain ontologies, a plurality of words associated with the third set of information via a machine learning module, wherein each of the plurality of words represents an analytical data corresponding to the real time contextual summary to be generated; generating, using each of the plurality of words identified, the real time contextual summary from the one or more analytical reports via the machine learning module, wherein the real time contextual summary is generated by implementing one or more Recurrent Neural Network (RNN) techniques; obtaining, from the plurality of words, a fourth set of information comprising one or more domain-specific sentences corresponding to the real time contextual summary to be generated by implementing the one or more RNN techniques; generating, from the fourth set of information, the real time contextual summary comprising statistical and non-statistical interpretations from the one or more analytical reports generated; identifying, based upon the third set of information, a sequence of representation of the fourth set of information to generate the real time contextual summary, wherein the sequence of representation is identified by implementing the one or more RNN techniques; and predicting an order of generation of each of the plurality of words by implementing the one or more RNN techniques for identifying the sequence of representation.
[007] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.

BRIEF DESCRIPTION OF THE DRAWINGS
[008] The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles.
[009] FIG. 1 illustrates a block diagram of a system for generating a real time contextual summary from one or more analytical reports by one or more machine learning techniques, in accordance with some embodiments of the present disclosure.
[010] FIG. 2 is an architectural diagram depicting components and flow of the system for generating the real time contextual summary from the one or more analytical reports by the one or more machine learning techniques, in accordance with some embodiments of the present disclosure.
[011] FIG. 3A through 3B is a flow diagram illustrating the steps involved in the process generating the real time contextual summary from the one or more analytical reports by the one or more machine learning techniques of in accordance with some embodiments of the present disclosure.

DETAILED DESCRIPTION OF EMBODIMENTS
[012] Exemplary embodiments are described with reference to the accompanying drawings. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the spirit and scope of the disclosed embodiments. It is intended that the following detailed description be considered as exemplary only, with the true scope and spirit being indicated by the following claims.
[013] Embodiments of the present disclosure provide for systems and methods for generating a real time contextual summary from one or more analytical reports by one or more machine learning techniques. Analytical processing comprises an approach to answer multi-dimensional analytical queries. Analytical Processing tools enable users to analyze multi-dimensional data by utilizing three basic analytical operations, like consolidation (aggregating data), drill-down (navigating details of data), and slice and dice (take specific sets of data and view from multiple viewpoints). Analytical processing systems comprise critical data structures allowing for fast analysis of data with the capability of manipulating and analyzing data from multiple perspectives. These facts and measures are commonly created from a star schema or a snowflake schema of tables in a Relational Database Management System (RDBMS).
[014] Analytical Processing comprises reporting, data exploration, data mining, data cleansing, information management, and business performance management. Analytical reporting tools create, maintain, or consume files such as documents, reports, dashboards, and the like. Analytical reports not only identify and analyze a problem and previously tried solutions, but provide of evidence-based recommendations to solve that problem where past attempts failed.
[015] However, an ability to act quickly and decisively in today's increasingly competitive marketplace is critical to the success of any organization. The volume of data that is available to organizations is rapidly increasing and frequently overwhelming. The availability of large volumes of data presents various challenges. One challenge is to avoid inundating an individual with unnecessary information. Another challenge is to ensure all relevant information is available in a real time manner.
[016] Traditional systems and methods such as data warehousing, decision support systems, online analytical processing systems etc. address the above mentioned challenges to some extent. However, the traditional systems and methods still suffer from various technical drawbacks. The traditional systems and methods require that the user connect via a computer system to the server system to initiate reports and view the contents of the reports. Moreover, the traditional systems and methods require that the user initiate a request for a report each time the user desires to have that report generated. A particular user may desire to run a particular report frequently to determine the status of the report.
[017] Further, reports may be extensive and may contain a large amount of information for a user to sort through each time a report is run. A particular user may only be interested in knowing if a particular value or set of values in the report has changed over a predetermined period of time. Finally, the traditional systems and methods require the user to initiate the new report and then scan through the new report to determine if the information has changed over the time period specified.
[018] Hence, there is a need for a methodology that provides for generating a contextual summary in real time from analytical outputs or analytical reports that provides for a real time data analysis and real time representation of critical information.
[019] Referring now to the drawings, and more particularly to FIG. 1 through 3B, where similar reference characters denote corresponding features consistently throughout the figures, there are shown preferred embodiments and these embodiments are described in the context of the following exemplary system and/or method.
[020] FIG. 1 illustrates an exemplary block diagram of a system 100 for generating a real time contextual summary from one or more analytical reports by one or more machine learning techniques, in accordance with an embodiment of the present disclosure. In an embodiment, the system 100 includes one or more processors 104, communication interface device(s) or input/output (I/O) interface(s) 106, and one or more data storage devices or memory 102 operatively coupled to the one or more processors 104. The one or more processors 104 that are hardware processors can be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions. Among other capabilities, the processor(s) is configured to fetch and execute computer-readable instructions stored in the memory 102. In an embodiment, the system 100 can be implemented in a variety of computing systems, such as laptop computers, notebooks, hand-held devices, workstations, mainframe computers, servers, a network cloud and the like.
[021] The I/O interface device(s) 106 can include a variety of software and hardware interfaces, for example, a web interface, a graphical user interface, and the like and can facilitate multiple communications within a wide variety of networks N/W and protocol types, including wired networks, for example, LAN, cable, etc., and wireless networks, such as WLAN, cellular, or satellite. In an embodiment, the I/O interface device(s) can include one or more ports for connecting a number of devices to one another or to another server.
[022] The memory 102 may include any computer-readable medium known in the art including, for example, volatile memory, such as static random access memory (SRAM) and dynamic random access memory (DRAM), and/or non-volatile memory, such as read only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes.
[023] According to an embodiment of the present disclosure, referring to FIG. 2, the architecture of the system 100 for generating the real time contextual summary from the one or more analytical reports by the one or more machine learning techniques may be considered in detail. An Automatic Speech Recognition (ASR) Engine 201 coverts voice-based inputs from user(s) into textual information using Deep Learning or other related techniques. One or more devices 202 comprise of a voice capturing device or any other device (for example, a keyboard) via which user inputs may be captured. A Query Builder 203 helps in building queries via a plurality of programming languages.
[024] A Database 204 facilitates capturing and analyzing all kinds of data associated with generating the real time contextual summary from the one or more analytical reports. A Text-to-Speech Engine 205 facilitates a conversion of analytical output into voice-enabled analytical output. One or more Domain Ontologies 206 comprise prebuilt ontologies by multiple words that are contextual to a plurality of technological or business domains. A Machine Learning Module 207 facilitates generating the real time contextual summary and also facilitates identification of words from the one or more Domain Ontologies 206.
[025] FIG. 3A through 3B, with reference to FIGS. 1 and 2, illustrates an exemplary flow diagram of a method of generating the real time contextual summary from the one or more analytical reports using the one or more machine learning techniques, in accordance with some embodiments of the present disclosure. In an embodiment the system 100 comprises one or more data storage devices of the memory 102 operatively coupled to the one or more hardware processors 104 and is configured to store instructions for execution of steps of the method by the one or more processors 104. The steps of the method of the present disclosure will now be explained with reference to the components of the system 100 as depicted in FIG. 1 and the flow diagram. In the embodiments of the present disclosure, the hardware processors 104 when configured the instructions performs one or more methodologies described herein.
[026] According to an embodiment of the present disclosure, at step 301, the one or more hardware processors 104 capture a first set of information comprising a set of user inputs via the one or more devices 202. The one or more devices 202 may comprise of a microphone, or any other voice capturing device or any other device via which user inputs may be captured, for example a keyword, or any combination thereof. In an embodiment, the system 100 may support a voice recognition software application (now shown in the figure). The voice recognition software comprises a software application (standalone) that can be integrated with word processing applications, e-mail applications, calendar applications, and so on.
[027] The voice recognition software operates with the use of a plurality of input devices capable of receiving a user's voice for input into the software application. The microphone may typically be used in conjunction with the voice recognition software for capturing the set of user inputs from a user. Further, the set of user inputs may be captured in the form of a text and / or as a voice-based input, for example, a natural language query. The set of user inputs may be captured in the form of one or more questions or in any other form. Considering an example scenario, the set of user inputs that may be captured may be as below:
How were actual vs. budgeted sales for organization X, please compare?
[028] According to an embodiment of the present disclosure, at step 302, the one or more hardware processors 104 extract, based upon the first set of information, a second set of information comprising user intents and entities corresponding to the first set of information via a cognitive engine (not shown in the figure), wherein the second set of information is extracted by implementing one or more natural language processing techniques. Initially, if the set of user inputs are captured as the voice-based input, the one or more hardware processors 104 convert, via the ASR Engine 201, voice-based user inputs into a set of textual information comprising of a plurality of text commands, wherein the conversion is performed using a Deep Neural Network (DNN) technique. The one or more hardware processors 104 initially store the set of voice-based user inputs into the memory 102 of the system 100 to get the set of voice-based user inputs communicated to the ASR Engine 201. The ASR Engine 201 is trained with training data using one or more DNN techniques.
[029] Further, the ASR Engine 201 (as implemented in the proposed methodology) leverages both trained acoustic model and language models. Acoustic model processing is used to model a received utterance. An output of the acoustic model is a lattice for each utterance. The lattice comprises an intermediate probability model in the form of a directed graph. The language model is used to calculate the probability of the whole transcript/sentence, not just the probability of the named entity. The language model during processing attempts to utilize the context information to compose phones from the candidate lattices to more meaningful word sequences. In case of the language model, lexicons are also considered so that all the nuances associated with voice based interaction of analytics application get gainfully trained.
[030] In an embodiment, upon capturing the set of user inputs, the one or more hardware processors 104 may implement an entity recognition process (or any other technique thereof) to detect entities mentioned in the set of user inputs captured, and links together multiple occurrences of same entity. Further, the one or more hardware processors 104 may also implement one or more fuzzy matching techniques (for example, a Levenshtein distance algorithm, a Soundex technique or any combination thereof) to generate one or more closely matching sentences with the set of user inputs captured.
[031] As is known in the art, the entity recognition process, inter-alia, performs a syntactic analysis, an entity detection of proper nouns and resolves one or more coreferences (that is, when two or more expressions in a text refer to the same person or thing in linguistics), and then determines to which named entities, the text in the closest matching sentence probably refers. These are referred to as recognized entities or recognized named entities. In an embodiment, different rules may be applied in the entity recognition for determining which antecedent is likely the best candidate entity.
[032] In an embodiment, the one or more natural language processing techniques may comprise for example, a Named Entity Recognition (NER) technique, for identifying a plurality of analytical entities by assigning analytical constructs with data artifacts based on the pre-defined analytical vocabulary, to extract the second set of information. As is known in the art, the plurality of analytical entities may comprise (but not limited to) types, measures, dimensions, and variants. As used herein, the term “dimension” refers to a structure that categorizes facts and measures. Examples of dimension include products, people, financial elements, time and the like. For example, a sales report may be viewed across the dimension of a product, a store, geography, a date, a quantity, revenue generated, and the like.
[033] A measure is a measurement data that may be manipulated, and usually denoted in some metric, for example, units, currency, etc. A variant can be a form or version that differs in some respect from other forms. In the example, “profit” is identified as a “measure” and “Asia region” is identified as a “variant.” In an example implementation, the second set of information may be extracted using the one or more natural language processing techniques by the cognitive engine as below:
Intent – “CompareActualBudgetedSales”
Entities – “Quarter 4” and “Line of Business and Region”
[034] According to an embodiment of the present disclosure, at step 303, the one or more hardware processors 104 generate a plurality of queries on the one or more analytical reports to be generated, wherein the plurality of queries are generated based upon the second set of information, and wherein the plurality of queries correspond to the first set of information (that is, the set of user inputs captured in the form of the one or more questions or via any other form). In an embodiment, the queries may be built by the one or more hardware processors 104 by implementing the query builder (203). Further, the pluralities of queries may be generated based upon the user intents and entities using any known techniques and methods, for example, a Markov modeling technique or a graph clustering technique.
[035] As is known in the art, the Markov modeling technique is employed to capture user’s behavior in manners that are different from previous random walk models for query log mining. For example, (1) it combines both the click-through and session co-occurrence information; and (2) it is an absorbing Markov model which makes limiting distributions of a random walk be dependent on the start node (which is a feature exploited for the current problem definition). As used herein, the term “queries” (from the expression “plurality of queries”) is referred to in the broadest sense to refer, to either two or more questions, one or more commands, or some form of input used as a control variable by the system, or any combination thereof. For example, a query may consist of a question directed to a particular topic, such as “what is a network” in the context of a remote learning application.
[036] In an embodiment, each of the plurality of queries are generated to obtain a plurality of datasets, wherein each of the plurality of datasets comprise analytical data corresponding to the one or more analytical reports to be generated. In an example implementation, the plurality of queries may be generated as below:
“Select actual Sales from tableName ‘Sales Budget Actual’ where (enterprise_name=’X’)”; and
“Select Budgeted Sales from tableName ‘Sales Budgeted’ where (enterprise_name=’X’ from Location_name=’Mumbai’ whose LineofBusiness=’Multimedia’)”.
[037] According to an embodiment of the present disclosure, at step 304, the one or more hardware processors 104 perform a plurality of steps in parallel using the plurality of queries. At step 304(i), the one or more hardware processors 104 generate the one or more analytical reports in a textual form corresponding to the plurality of user intents and entities. Thus, the one or more analytical reports (from which the real time contextual summary is to be generated) are generated in a textual form in this step based upon each of the plurality of queries.
[038] In an embodiment, the one or more analytical reports comprise analytical data for which the real time contextual summary is to be generated. The one or more hardware processors 104 initially execute the plurality of queries to obtain one or more data points for generating the one or more analytical reports. As is known in the art, a data point is a smallest individual entity on a chart. On non-Shape charts, the one or more data points are represented depending upon their chart type. For example, a line series consists of one or more connected data points. On Shape charts, the one or more data points are represented by individual slices or segments that add up to the whole chart. For example, on a pie chart, each piece is a data point. The one or more data points form a series. By default, all formatting options may be applied to each of the one or more data points in the series.
[039] At step 304(ii), the one or more hardware processors 104 determine a third set of information comprising statistical data corresponding to the one or more analytical reports to be generated. The proposed disclosure implements one or more advanced statistical techniques, for example, a correlation analysis, and the like along with one or more known simple analysis techniques, for example, standard deviation, to analyze datasets in each of the plurality of queries. By implementing statistical techniques mentioned above, the one or more hardware processors 104 generate a correlation amongst values of various columns in the plurality of datasets.
[040] Further, the correlation analysis facilitates identification a correlation amongst one or more dependent variables and one or more independent variables in the plurality of datasets. The identification results in further identification of a plurality of trends and patterns in the plurality of datasets. Considering same example scenario as in previous steps, the third set of information may be generated as shown in Table 1 below:
Table 1
Product Id Actual Sales Budgeted Sales Quarter
H740 400450 500000 Q12017
P3000 1250000 1000000 Q12017
Router 1550000 1500000 Q12017
Home Automation 1022000 1000000 Q12017
[041] According to an embodiment of the present disclosure, at step 305, the one or more hardware processors 104 identify, from the one or more domain ontologies 206, a plurality of words associated with the third set of information (comprising the statistical data) by implementing the machine learning module 207, wherein each of the plurality of words represents an analytical data corresponding to the real time contextual summary to be generated. The one or more domain ontologies 206 comprise prebuilt ontologies using words that are contextual to a plurality of technological or business domains.
[042] In general, an ontology may include a linked collection of words, concepts, or issues that relate to the environment or domain that pertains to the phrase, sentence, paragraph, or discourse. The ontology may also define relationships among a number of language elements that are likely to be used, for example, in a particular context, and/or are related to particular subject matter. Considering an example scenario, the one or more domain ontologies 206 may comprise (but not limited to) the plurality of words as “Highest, Selling, Product, Budgeted, Actual, Lowest, Gross, Margin, Variance, Positive, Variance, Quarter” etc. and any combination of one or more of the plurality of words.
[043] In an embodiment, each of the plurality of words associated with the statistical data may identified from the one or more domain ontologies 206 by implementing an N-grams technique. As is known in the art, the N-grams technique allows machine learning algorithms such as support vector machines to learn from string data. N-grams comprises of one or more word prediction algorithm using probabilistic methods to predict next word after observing N-1 words. Therefore, computing the probability of the next word is closely related to computing the probability of a sequence of words. Considering an example scenario, for word “TEXT”, N-grams may be generated as follows:
bi-grams: _T, TE, EX, XT, T_ tri-grams: _TE, TEX, EXT, XT_, T_ _ quad-grams: _TEX, TEXT, EXT_, XT_ _, T_ _ _
[044] In an embodiment, the N-grams technique may be implemented using python or any other natural language for performing the classification. In an example implementation of step 305, suppose the third set of information generated comprises “All products met the budgeted sales target except one product i.e. H740”, and the real time contextual summary to be generated needs to have information for a highest selling product, the N-grams technique scans for a set of words that may immediately succeed the word “highest”. Scanning may result in words like “rate” or “margin” or sequence of words like “selling product”. The machine learning module 207 selects “selling product” as the third set of information (or the plurality of queries) do not comprise the words “rate” or “margin”.
[045] According to an embodiment of the present disclosure, at step 306, the one or more hardware processors 104 generate, using each of the plurality of words identified, the real time contextual summary from the one or more analytical reports (generated in step 304(i)) via the machine learning module 207 by implementing one or more Recurrent Neural Network (RNN) techniques. None of the traditional systems and methods provide for generating real time contextual summaries or real time analytical summaries of analytical reports by implementing one or more machine learning or RNN techniques. The proposed disclosure thus facilitates generating the real time contextual summary from the one or more analytical reports dynamically from the plurality of words, that is, based upon the statistical data and the plurality of datasets. The process of generating the real time contextual summary may now be considered in detail.
[046] Generally, neural networks are machine learning models that employ one or more layers of nonlinear units to predict an output for a received input. Some neural networks include one or more hidden layers in addition to an output layer. The output of each hidden layer is used as input to the next layer in the network, i.e., the next hidden layer or the output layer. Each layer of the network generates an output from a received input in accordance with current values of a respective set of parameters.
[047] Further, some neural networks, for example, those that are designed for time series problems or sequence-to-sequence learning incorporate recurrent loops which permit memory, in the form of a hidden state variable, to persist within a layer between data inputs. These comprise RNNs. In a "recurrent" neural network architecture, inputs are received by a single layer of neurons and the activations of those neurons are fed back as inputs to that single layer to produce new activations during a "propagation".
[048] In an embodiment, at step 306, the one or more hardware processors 104 initially obtain, from the plurality of words, a fourth set of information comprising one or more domain-specific sentences corresponding to the real time contextual summary to be generated. For obtaining the fourth set of information, the machine learning module 207 may implement a long short-term memory (LSTM) neural network technique. A variation of RNNs, LSTM neural networks include multiple gates within each layer to control the persistence of data between data inputs.
[049] In an embodiment, the RNN or the LSTM technique works by taking a word amongst the plurality of words , its context from previous time steps (as mediated by the hidden layers) and predicts the next character in the sequence. Each of the plurality of words comprise a plurality of letter vectors that are trained with via backpropagation technique. The plurality of letter vectors may then be combined through a (learnable) matrix-vector multiply transformation technique into a first hidden layer representation (not shown in the figure), then into a second hidden layer representation (not shown in the figure), and finally into an output space (not shown in the figure). The output space comprises a dimensionality equal to the number of characters in the dataset and every dimension provides a probability of the next character in the sequence. LSTM network is therefore trained to always predict the next character or word by performing a softmax function and a cross-entropy loss on all letters).
[050] Thus, upon determining the third set of information, the one or more RNN techniques may be applied on each of the plurality of words, that is, on each character forming the third set of information to identify a sequence of representation of the fourth set of information. The sequence of representation thus facilitates a prediction of an order of generation of each of the plurality of words. The one or more RNN techniques thus predict the next word by learning context from one or more previous words. Considering an example scenario, the fourth set of information may be generated as (but not limited to):
"Highest selling product in 2017 is Router"; and
"Highest Gross Margin in Q1 2017 is for Router"
[051] In an embodiment, the one or more hardware processors 104 generate the real time contextual summary from the fourth set of information via the machine learning module 207, wherein the real time contextual summary comprises statistical and non-statistical interpretations (or any other interpretations) on the one or more analytical reports generated.
[052] According to an embodiment of the present disclosure, advantages of the proposed disclosure may now be considered in detail. As explained via steps 301 through 306, the proposed disclosure facilitates a real time summary generation of analytical results. This saves lot of time and manual efforts involved in assessing analytical reports, as the real time contextual summary generated by the proposed methodology comprises a filtered set of critical / meaningful / in-depth information, which is filtered from the analytical reports itself. None of the traditional systems and methods provide for a dynamic or real time contextual summary generation via machine learning or other techniques.
[053] By providing for the real time contextual summary generation, the proposed disclosure facilitates a real time decision making. By generating the real time contextual summary, the proposed disclosure also eliminates unnecessary data or information that may finally not be required by an organization. The proposed disclosure facilitates the real time contextual summary generation from even complex of high-dimensional data types or datasets. Finally, the proposed methodology does not require any separate templates to generate the real time contextual summary as the real time contextual summary may be generated directly from the one or more analytical reports by implementing the one or more machine learning techniques.
[054] In an embodiment, the memory 102 can be configured to store any data that is associated with generating the real time contextual summary from the one or more analytical reports by the one or more machine learning techniques. In an embodiment, the information pertaining to the first set of information the second set of information, the third set of information, the fourth set of information, the plurality of queries, and the real time contextual summary generated etc. is stored in the memory 102. Further, all information (inputs, outputs and so on) pertaining to generating the real time contextual summary from the one or more analytical reports by the one or more machine learning techniques may also be stored in a database (other than the database 204), as history data, for reference purpose.
[055] The written description describes the subject matter herein to enable any person skilled in the art to make and use the embodiments. The scope of the subject matter embodiments is defined by the claims and may include other modifications that occur to those skilled in the art. Such other modifications are intended to be within the scope of the claims if they have similar elements that do not differ from the literal language of the claims or if they include equivalent elements with insubstantial differences from the literal language of the claims.
[056] It is to be understood that the scope of the protection is extended to such a program and in addition to a computer-readable means having a message therein; such computer-readable storage means contain program-code means for implementation of one or more steps of the method, when the program runs on a server or mobile device or any suitable programmable device. The hardware device can be any kind of device which can be programmed including e.g. any kind of computer like a server or a personal computer, or the like, or any combination thereof. The device may also include means which could be e.g. hardware means like e.g. an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a combination of hardware and software means, e.g. an ASIC and an FPGA, or at least one microprocessor and at least one memory with software modules located therein. Thus, the means can include both hardware means and software means. The method embodiments described herein could be implemented in hardware and software. The device may also include software means. Alternatively, the embodiments may be implemented on different hardware devices, e.g. using a plurality of CPUs.
[057] The embodiments herein can comprise hardware and software elements. The embodiments that are implemented in software include but are not limited to, firmware, resident software, microcode, etc. The functions performed by various modules described herein may be implemented in other modules or combinations of other modules. For the purposes of this description, a computer-usable or computer readable medium can be any apparatus that can comprise, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
[058] The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed. These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed. Alternatives (including equivalents, extensions, variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope and spirit of the disclosed embodiments. Also, the words “comprising,” “having,” “containing,” and “including,” and other similar forms are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items. It must also be noted that as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise.
[059] Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term “computer-readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., be non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.
[060] It is intended that the disclosure and examples be considered as exemplary only, with a true scope and spirit of disclosed embodiments being indicated by the following claims.