Optum Technology, Inc.

Optum, Inc.

Infosys Limited

Optum Services (Ireland) Limited

Infosys Technologies Limited

Suman Roy

Ayan Sengupta

Vijay Varma Malladi

Amit Kumar

Gaurav Ranjan

Shashi Kumar

Michael Bridges

Tanmoy Chakraborty

William Scott Paka

Siddhartha Asthana

Kiran Prakash Sawant

Ashish Chaturvedi

Ahmed Selim

Riccardo Mattivi

Irfan Bulu

Thomas G Sullivan

Ming Fei Yan

Vishal Pathak

Anurag Miglani

Gunjan Balde

Malladi Vijay Varma

Sumit Jindal

Lia O Solis Obineche

Surajit Das

Matthew J Stewart

Sourabh Kumar Bhattacharjee

Souparna Das

Olivier Charvin

Rahul Chetlangia

Prakhar Pratap

Madhvi Gupta

Kaustav Mukherjee

Abraham Gebru Tesfay

Various embodiments of the present disclosure provide summarization techniques for summarizing complex documents, such as long unstructured call transcripts. The summarization techniques include generating a plurality of interaction topics for an interaction transcript and iteratively summarizing each interaction topic based on a preceding partial summary for the interaction transcript that corresponds to a preceding interaction topic that precedes the interaction topic in the interaction transcript. An abstractive summary is generated using a recursive abstractive model that is trained using training data generated based on holistic similarity scores between interaction topics of a call transcript and summary sentences of a corresponding target summary.

Machine learning divide and conquer techniques for long dialog summarization

Various embodiments provide methods, apparatus, systems, computing entities, and/or the like, for providing a summarization of a conversation, such as a telephonic conversation. In an embodiment, a method is provided. The method comprises receiving an input data object comprising textual data of a conversation, the textual data comprising sentence-level tokens. The method further comprises classifying some sentence-level tokens as interrogative sentence-level tokens, and identifying subtopic portions of the textual data, each interrogative sentence-level token located within one subtopic portion. The method further comprises determining whether an interrogative sentence-level token is substantially similar to one of a plurality of target queries, and for such interrogative sentence-level tokens, selecting sentence-level tokens from a subtopic portion corresponding to the such interrogative sentence-level tokens. The method then comprises generating a summarization data object comprising the selected sentence-level tokens for each interrogative sentence-level token substantially similar to a target query and performing summarization-based actions.

Query-focused extractive text summarization of textual data

Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing natural language processing operations using an attention-based text encoder machine learning model that is trained using a multi-task training routine that is associated with two or more training tasks (e.g., a multi-task training routine that is associated with two or more sequential training tasks, a multi-training routine that is associated with two or more concurrent training tasks, and/or the like).

Natural language processing machine learning frameworks trained using multi-task training routines

Various embodiments provide for summarization of an interaction, conversation, encounter, and/or the like in at least an abstractive manner. In one example embodiment, a method is provided. The method includes generating, using an encoder-decoder machine learning model, a party-agnostic representation data object for each utterance data object. The method further includes generating an attention graph data object to represent semantic and party-wise relationships between a plurality of utterance data objects. The method further includes modifying, using the attention graph data object, the party-agnostic representation data object for each utterance data object to form a party-wise representation data object for each utterance data object. The method further includes selecting a subset of party-wise representation data objects for each of a plurality of parties. The method further includes decoding, using the encoder-decoder machine learning model, the subset of party-wise representation data objects for each party to form abstractive summary data object(s).

Supervised and unsupervised machine learning techniques for communication summarization

Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing natural language processing operations using a hybrid reason code prediction machine learning framework. Certain embodiments of the present invention utilize systems, methods, and computer program products that perform natural language processing using a hybrid reason code prediction machine learning framework that comprises one or more of the following: (i) a hierarchical transformer machine learning model, (ii) an utterance prediction machine learning model, (iii) an attention distribution generation machine learning model, (iv) an utterance-code pair prediction machine learning model, and (v) a hybrid prediction machine learning model.

Natural language processing techniques using hybrid reason code prediction machine learning frameworks

There is a need for more effective and efficient text categorization. This need can be addressed by, for example, techniques for semantic text categorization. In one example, a method includes determining an input vector-based representation of an input document; processing the input vector-based representation using a trained supervised machine learning model to generate the categorization based at least in part on the input vector-based representation, wherein: (i) the trained supervised machine learning model has been trained using automatically-generated training data, and (ii) the automatically generated training data is generated by determining an inferred semantic label for each unlabeled training document of one or more unlabeled training documents; and performing one or more categorization-based actions based at least in part on the categorization, and (iii) the labels are described by one or more short documents/short texts.

Unsupervised approach to assignment of pre-defined labels to text documents

There is a need for more effective and efficient predictive natural language topic detection. This need can be addressed by, for example, solutions for performing sequential topic detection. In one example, a method includes determining a sequential topic distribution data object for the current document sequence, determining a current term-context correlation data object for the current document sequence, determining a current context-topic correlation data object for the current document sequence, determining an updated term-topic correlation data object based at least in part on the current context-topic correlation data object, determining topic modeling predictions based at least in part on the sequential topic distribution data object and the updated term-topic correlation data object, and performing prediction-based actions based at least in part on the topic modeling predictions.

Natural language processing techniques for sequential topic modeling

Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing text similarity determination. Certain embodiments of the present invention utilize systems, methods, and computer program products that perform text similarity determination by using at least one of Word Mover's Similarity measures, Relaxed Word Mover's Similarity measures, and Related Relaxed Word Mover's Similarity measures.

Machine learning techniques for word-based text similarity determinations

There is a need for more effective and efficient natural language processing (NLP) solutions. This need can be addressed by, for example, solutions for performing NLP-based document prioritization by utilizing joint sentiment-topic (JST) modeling. In one example, a method comprises identifying a JST latent distribution of the digital document that describes topic designation probabilities and sentiment designation probabilities for the digital document; determining, by processing the topic designation probabilities, a document-topic entropy measure for the digital document; determining, by processing the sentiment designation probabilities, a sentiment-topic entropy measure for the digital document; determining, by processing per-word inverse domain frequency measures for the digital, a document popularity measure for the digital document; generating the predicted document priority score based on the document-topic entropy measure, the sentiment-topic entropy measure, and the document popularity measure; and performing one or more prediction-based actions based on the predicted document priority score.

Natural language processing techniques using joint sentiment-topic modeling

As described herein, various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing natural language processing operations using a combination of a cross-token attention machine learning, a cross-utterance attention machine learning model, and an integer linear programming joint keyword-utterance optimization model to select an extractive keyword summarization of a multi-party communication transcript data object that comprises a selected utterance subset of U utterances (e.g., U sentences) of a document data object and a selected keyword subset of K candidate keywords of the document data object.

Natural language processing techniques for document summarization using local and corpus-wide inferences

There is a need for more effective and efficient predictive natural language summarization. This need can be addressed by, for example, solutions for performing predictive natural language summarization using a constrained optimization model. In one example, a method includes identifying one or more per-party utterance subsets in a multi-party call transcript; generating a plurality of eligible extractive summaries that comply with one or more optimization constraints; for each eligible extractive summary of the plurality of eligible extractive summaries, determining an overall summary utility measure; generating the optimal extractive summary based at least in part on each overall summary utility measure for an eligible extractive summary of the plurality of eligible extractive summaries; and performing one or more summary-based actions based at least in part on the optimal extractive summary.

Natural language processing for optimized extractive summarization

Various embodiments of the present invention provide methods, apparatus, systems, computing devices, computing entities, and/or the like for performing predictive structural analysis on document data objects that are associated with an ontology graph. Certain embodiments of the present invention utilize systems, methods, and computer program products that perform predictive data analysis operations on document data objects that are associated with an ontology graph using document embeddings that are generated by graph-embedding-based paragraph vector machine learning models.

Graph-embedding-based paragraph vector machine learning models

Growing community of inventors

Bengaluru, India

Suman Roy