Accenture Global Solutions Limited

Nirav Sampat

Prakash Ghatage

Naveen Kumar Thangaraj

Vinu Varghese

Kumar Viswanathan

Saran Prasad

Balaji Janarthanam

Anil Kv Kumar

Kaustubh Kurhekar

Selvakuberan Karuppasamy

Sattish Sundarakrishnan

Shikhar Srivastava

Abhishek Patni

Hari Kumar Karnati

Harsh Kumar Katiyar

Madhura Shivaram

Sriram Lakshminarasimhan

Guruprasad Dasappa

Nagendra B Kumar

Suraj Govind Jadhav

Manish Jain

Ullas Balan Nambiar

Ashish Jain

Sumeet Sawarkar

Siddesha Swamy

Rahul Kotnala

Kaushal Mody

Anil Kumar

Karthik Meenakshisundaram

Santhosh Kumar Rajavelu Balachandar

Samatha Kottha

Sreevidya Prasad

Ravi Kant Gaur

Amit Porwal

Narayan Sethi

Billy Hart

Bhushan Ananda Patil

Elaine T Bombal

Prabhat Kumar Singh

Suvendu Kumar Mahapatra

Srikanth Narayanan

Rekha Mani

Dharmesh Dhirajlal Barochia

Aravind Krishnan

Purnanga Prema Borah

Kameshkumar Lakshminarayanan

Deepali Jain

Peter S Carpenter

Suhas Vinod Sane

Anto Jose

Karthik Srikantamurthy

Saumyabrata Mitra

Ragav Devanathan

Richard Stephen Vincent Price

Christina Catharina De Vries

Dheeraj Kharya

Madhu Sudhan H V

Amit Vithoba Patil

Kunal Jaiwant Kharsadia

Pavel Sem

Arun Viswanath

Deepam Biswas

Vinay Verma

Ujjala Chattopadhyay

Basaveswara Kiran Dumpala

Tomas Chaloupka

An Artificial Intelligence (AI) Driven multi-platform, multi-lingual translation system analyzes the speech context of each audio stream in the received audio input, selects one of a plurality translation engines based on the speech context, and provides translated audio output. If audio input from multiple speakers is provided in a single channel then it is diarized into multiple channels so that each speaker transmitted on a corresponding channel to improve audio quality. A translated textual output received from the selected translation engine is modified with sentiment data and converted into an audio format to be provided as the audio output.

Multi-platform voice analysis and translation

A device may process data associated with a first entity and a second entity, using one or more machine learning techniques, to identify a set of trends associated with a set of documents. The device may receive new documents associated with the first entity and the second entity. The device may generate, based on the set of trends, a first set of exceptions indicating that a new document is associated with a first type of error. The device may generate, using a similarity analysis technique, a second set of exceptions indicating that a new document is associated with a second type of error. The device may communicate with one or more systems to perform one or more actions associated with correction or prevention of processing errors relating to the new document based a claim relating to the first set of exceptions or the second set of exceptions.

End-to-end identification of erroneous data using machine learning and similarity analysis

A system for obtaining optimized regular expression may convert a received input data into a plurality of embeddings. The system may receive a generated regular expression corresponding to the plurality of embeddings, wherein the generated regular expression is at least one of an existing regular expression from a database and a newly generated regular expression. The system may parse the generated regular expression into a plurality of sub-blocks. The system may classify the plurality of sub-blocks to obtain a plurality of classified sub-blocks. The system may evaluate a quantifier class for each classified sub-block to identify a corresponding computationally expensive class. The system may perform an iterative analysis to obtain a plurality of optimized sub-blocks associated with a minimum computation time. The system may combine the plurality of optimized sub-blocks to obtain the optimized regular expression.

Automated generation and integration of an optimized regular expression

Systems and methods for detecting, classifying, and managing impediments are disclosed. For example, embodiments may be related to impediments in project management. The proposed systems and methods are configured to evaluate data harvested from multiple different sources (in different formats), identify potential impediments that may be described or present in the data, and classify said impediments based on whether the impediment is non-technical or technical. In addition, the proposed systems implement a technical solution of active learning combined with reinforcement learning to produce a feedback loop that, over each iteration, improves the accuracy of the impediment classification. The impediment management assistant is configured to identify impediments from various inputs sources across industries with an AI-based self-learning capability, providing a robust and accurate model even with only a limited training dataset.

Detection and classification of impediments

Systems and methods for managing supply chain of products and services are disclosed herein. A system generates supply chain data based on historical data received from data sources corresponding to supply chain of product or service. Further, system extracts data entity and set of attributes from supply chain data, to determine semantically related data entities. Furthermore, system determines use case corresponding to management of supply chain, based on semantically related data entities. Additionally, system predicts, risk or priority associated with product or service in the supply chain, to generate risks and alerts, based on prediction. Further, system assigns critical and high-priority use case to one or more agents based on a performance score of the one or more agents. Furthermore, system provides insights and suggestions for managing the supply chain of product or service at regional level and global level of supply chain.

System and method for assigning agents to critical/high priority supply chain risk use cases

An authentication and applications access system provides access to a plurality of backend applications via a social media application installed on a user device. Metadata associated with a request to access one of the plurality of backend applications is initially extracted. The metadata is used to authenticate the request via a plurality of validation steps. Upon authentication, the information from the request is provided to the backend application to receive the results responsive to the request. Any sensitive data included in the results is suppressed via data substitution steps from the transmission to the user device. The output from the applications access system with the sensitive data occluded is provided for display on a social media user interface (UI) on a user device.

Enabling secure access to applications via social media

A device may receive log data from application logs associated with applications, service logs associated with services, and server logs associated with server devices. The device may store the log data. The device may perform natural language processing on the log data to convert the log data into event data identifying events associated with categories. The device may process the event data, with a first machine learning model, to identify patterns in the event data and to generate an alert based on the patterns. The device may process the event data, with a second machine learning model, to generate a correlation matrix for the event data and to predict an event based on the correlation matrix. The device may process the event data, with a third machine learning model, to classify the event data based on the categories and to generate a recommendation based on classifying the event data.

Utilizing machine learning models with a centralized repository of log data to predict events and generate alerts and recommendations

An Artificial Intelligence (AI) based document digitization, transformation and validation system extracts fields from digital documents via different document digitization processes. A document packet with a plurality of documents is initially accessed and any non-digital documents in the document packet are digitized. The errors in the digitized documents are corrected and non-English documents are translated into English. Each of the documents is provided to a plurality of digitization services for the extraction of fields by a plurality of field extraction models. If a field has multiple field instances extracted by more than one digitization service, then a field instance with the highest confidence score is selected for inclusion into the consolidated results. The consolidated results produced in different JavaScript Object Notation (JSON) formats are converted into a common JSON format which may be further validated and provided to downstream processes.

Document digitization, transformation and validation

A system for solution architecture prediction may identify a previous solution from a data source and may create a historical solution evaluation matrix by mapping a plurality of concern categories with the previous solution. The system may identify a plurality of solution components preponderant to deriving a solution associated with the solution architecture prediction and create a potential solution evaluation matrix therefrom. The system may evaluate the historical solution evaluation matrix and the potential solution evaluation matrix to determine a credibility score for each solution comprised therein. Based on the evaluation, a solution prediction data may be generated including a previous solution, a potential solution, and the associated credibility score. A service solution may be selected from the solution prediction data to resolve the solution prediction requirement.

Solution architecture prediction

A device may receive input data identifying a technical architecture diagram, a design document, an interface specification document, and technical architecture icons, and may process the input data identifying the technical architecture diagram, with a model, to determine hierarchical objects from the technical architecture diagram. The device may perform OCR and NLP of the hierarchical objects to determine blocks of data, and may compare the blocks of data and the input data identifying the design document to identify functionalities of applications. The device may compare the blocks of data and the input data identifying the interface specification document to identify attributes, and may compare the blocks of data and the input data identifying the technical architecture icons to identify icons. The device may consolidate the blocks of data, the functionalities, the attributes, and the icons into a final document, and may perform actions based on the final document.

Utilizing deep learning and natural language processing to convert a technical architecture diagram into an interactive technical architecture diagram

Systems and methods for email validation are disclosed. The email validation includes transforming format of emails to a predefined format understandable the present system and application of text mining component on the transformed format. The email validation further includes obtaining details from a repository related to a historical pattern associated with an email validation requirement and a cognitive learning operation employed for the historical email validation to ascertain an outcome of the historical validation for similar emails. The email validation also includes predicting misdirection of the email and change in configuration of the email account based on the validation of the email.

Email validation

An example chatbot generation platform may receive a request to generate a chatbot; determine a chatbot template for the chatbot based on the request; obtain custom chatbot information according to the chatbot template; generate a chatbot corpus for the chatbot using the custom chatbot information and the chatbot template; generate a set of question and answer (QnA) pairs based on the chatbot corpus; configure a language analysis model for the chatbot; build the chatbot according to the set of QnA pairs and the language analysis model; and deploy the chatbot to a chatbot host platform for operation. The chatbot may be built to engage in an interaction with a user via the chatbot host platform, use the language analysis model to select one or more QnA pairs from the set of QnA pairs during the interaction, and train the language analysis model based on the interaction.

Chatbot generator platform

A data narration generating system generates snippets that include representations of data in one of a plurality of formats for inclusion into a data narration. The narration generating system receives selected data to be included into the data narration and provides the selected data to a plurality of ML models. The plurality of ML models are trained in generating snippets in one of the plurality of formats which can include textual format and a tabular format. Snippets in graphical formats can also be generated by rule-based processes. A plurality of snippets are thus generated in one or more of the plurality of formats which can then be presented to a user for selection and inclusion into the data narration. Alternately, a subset of the plurality of snippets can also be selected automatically based on a quality and quantity of data and a voting mechanism. The data narration thus generated is further configured to present different views based on privileges associated with user profiles.

Artificial intelligence (AI) based generation of data presentations

A device may receive, from a user device, a request to access a virtual reality (VR) learning environment that includes an identifier associated with a program that supports the VR learning environment. The device may identify a set of objects to use within the VR learning environment by searching a data structure using the identifier associated with the program that supports the VR learning environment. The device may provide the VR learning environment to the user device. The device may receive, from the user device, information associated with interactions within the VR learning environment. The device may identify one or more additional objects to use within the VR learning environment by using one or more natural language processing techniques to analyze the information associated with the interactions within the VR learning environment. The device may provide the one or more additional objects to the user device.

Customized virtual reality learning environment

Techniques are described for routing service requests in a computer-implemented service environment. A received service request may be initially analyzed to determine a priority of the request. In some implementations, one or more actions may be automatically performed to provide an initial response to the requester. The text of the request may be analyzed to automatically determine a category of the request. In some implementations, a classification engine may determine the category of the request through use of a classification model that has been trained using one or more machine learning (ML) techniques and/or that employs Natural Language Processing (NLP). Based on the category, the request may be routed to agent(s) for handling. Routing may include generating a ticket that includes the request, the category, the priority, and/or other information, and the ticket may be provided to the appropriate agent(s) through a ticketing service.

Learning based routing of service requests

Growing community of inventors

Mumbai, India

Nirav Sampat