Infosys Limited

Zuora, Inc.

Live Objects, Inc.

Sudipto Shankar Dasgupta

Mayoor Rao

Kamesh Raghavendra

Michael Reh

Ganapathy Subramanian

Swaminathan Natarajan

Sairam Yeturi

Navin Budhiraja

Arulkumar Gopalan

John Kuriakose

Prasanna Nagesh Teli

Sameer Mahadeo Kolhatkar

Gopinath Srungarapu

Kiran Kumar Kaipa

Vivek Sinha

Abdul Razack

A computing server may receive master data, transaction data, and one or more existing process models of a domain. The computing server may aggregate, based on domain knowledge ontology of the domain, the master data and the transaction data to generate a fact table. For example, entries in the fact table may be identified as relevant to the target process model and include attributes and facts that are extracted from master data or transaction data. The computing server may convert the entries in the fact table into vectors. The computing server inputting vectors into one or more machine learning algorithms to generate one or more algorithm outputs. One or more algorithm outputs may correspond to one or more improved process models that are optimized compared to the existing process models. The computing server may provide the improved process model to the domain to replace one of the existing process models.

Dynamic process model optimization in domains

A computing server configured to process data of a domain from unstructured data sources to generate natural language phrases describing relationships between entities identified from the unstructured data. The computing server may receive master data schema and domain knowledge ontology of a domain including relationship definitions in the domain. The computing server may identify targeted types of named entities of the domain from the master data schema according to the relationship definitions in the domain knowledge ontology. The computing server may extract a plurality of named entities from unstructured data of the domain. The computing server may generate one or more sequences of named entities and assign entity labels to the named entities. The computing server may, based on the entity labels, generate natural language phrases describing relationships of sets of named entities.

Dynamically trained models of named entity recognition over unstructured data

A computing server configured to process data of a domain from heterogeneous data sources. A domain may store data and schema, domain knowledge ontology such as resource description framework, and unstructured data. The computing server may extract objects from the unstructured data. The computing server may convert the extracted named entities and activities to word embeddings and input the word embeddings to a machine learning model to generate an activity time sequence. The machine learning model may be a long short-term memory. A process model may be generated from the time sequence. The computing server may identify outliers in the process model based on metrics defined by the domain. The computing server may convert transactions without outliers as word embeddings and generate signatures of the transactions using cosine similarity. The computing server may augment the results with the domain knowledge ontology.

Generation of process models in domains with unstructured data

A computing server may receive master data, transaction data, and a process model of a domain. The computing server may aggregate, based on domain knowledge ontology of the domain, the master data and the transaction data to generate a fact table. For example, entries in the fact table may be identified as relevant to the target process model and include attributes and facts that are extracted from master data or transaction data. The computing server may convert the entries in the fact table into vectors. The computing server may identify, based on the vectors, an attribute in the process model as being statistically significant on impacting the process model. For example, a regression model may be used to determine the statistical significance of an attribute on the model process. The computing server may generate an action associated with the attribute to improve the process model.

Automated process collaboration platform in domains

A system and method of creating an entity relationship map includes receiving a stream of lexical matter associated with one or more categories () and identifying one or more tokens from the received lexical matter based on the one or more categories (). A frequency of one or more of unique lexical token and recurring lexical token are determined () and one or more outliers based on a standard deviation range associated with the at least one category is eliminated (). Sentences with the one or more recurring lexical tokens are selected () to find one or more lexical neighbors and the entity relationship map is created based on an association between the unique lexical tokens and the at least one lexical neighbor ().

Extraction of tokens and relationship between tokens from documents to form an entity relationship map

A method and system for authenticating software licenses of a software includes a request for a software authentication received from one or more software subscribers and one or more electronic licenses distributed between one or more software vendors and the one or more software subscribers. Further, one or more tokens are validated through an authentication engine at a delivery packet delivered to the software subscriber. A license key associated with each validated token is generated and distributed through a licensing engine. The software is initiated to be enabled through the license key.

Cryptographic mechanisms for software setup using token-based two-factor authentication

A method and system can implement error and event log correlation in an apparatus and include extracting one or more log information associated with a storage location and creating a flexible structure of the one or more log information. The one or more log information is translated to a database store based on a user input. A match level is determined between an event and error data through the one or more log information extracted. When the match level exceeds a predetermined value, a relationship between the event and error data is created through an algorithm and a shareable entry is created for the relationship in a format usable by another apparatus.

Method and system of automatic event and error correlation from log data

A method and/or system for heterogeneous predictive models generation based on sampling of big data is disclosed. The method involves receiving a dataset and a target column associated with the dataset at a data processing engine from a distributed data warehouse. One or more columns associated with the dataset are classified at the data processing engine as a categorical column or a continuous column. One or more parameters in the dataset are identified to extract a sample data from the dataset. The sample data from the dataset is extracted based on the identified one or more parameters. One or more rank ordered machine learning algorithms are recommended to one or more users, to generate one or more predictive models from the sample data. One or more heterogeneous predictive models are generated based on the rank ordered algorithm through one or more iterations.

Automated system for development and deployment of heterogeneous predictive models

A method generating a platform-agnostic abstract syntax tree (AST) comprises receiving data in a predefined format, through an input unit; subsequently parsing the data to extract model information corresponding to the predefined format of the data; and transforming, by a processing server, the model information to an abstract syntax tree (AST) structure. The above steps aid in generating, by the processing server, a platform-agnostic AST by combining predefined metadata and the abstract syntax tree (AST) structure.

System and method of generating platform-agnostic abstract syntax tree

A method and system of a data join includes capture of metadata information associated with one of semi-structured data and unstructured data. A flattened structure for one of the semi-structured data and the unstructured data is defined, and an entity is extracted from the unstructured data. Further, one of the semi-structured data and an entity extracted unstructured data are flattened based on the flattened structure, and flattened semi-structured data and flattened entity extracted unstructured data with relational data are joined.

System and method of data join and metadata configuration

A method and system automates training named entity recognition in natural language processing to build configurable entity definitions includes receiving input documents or entities through an administration module and defining a domain for each entity. Further, one or more entities corresponding to the domain specific entity in the received documents are determined and a training file to one of pick a right parser, extract content and label the entity ambiguity is generated. One or more user actions are collected and maintained at a repository through a knowledge engine. Still further, one or more labelled ambiguous words are predicted and the knowledge engine is updated. Data may be fetched, through a training pipeline execution engine and each entity may be associated with one or more documents based on the fetched data from the document store to build configurable entity definitions.

Growing community of inventors

Sunnyvale, CA, United States of America

Sudipto Shankar Dasgupta