International Business Machines Corporation

Nuance Communications, Inc.

Kyndryl, Inc.

Sachindra Joshi

Ashish Verma

Shantanu Ravindra Godbole

Jitendra Ajmera

Dinesh Raghu

Ajay Gupta

Raghuram Krishnapuram

Danish Contractor

Indrajit Bhattacharya

Jatin Ganhotra

Kevin W English

Ashwin Srinivasan

Jeffrey Mark Achtermann

Chulaka Gunasekara

Luis Alfonso Lastras-Montano

Mukesh Kumar Mohania

Pankaj Dhoolia

Sathish R Indurthi

Kenney Ng

Sumit Negi

Srujana Merugu

Prasanta Kumar Ghosh

Guy Feigenblat

Harshit Kumar

Om Dadaji Deshmukh

Benjamin Sznajder

Nathaniel Mills

Neeraj Agrawal

David Nahamoo

Sameep Mehta

L Venkata Subramaniam

Pratyush Kumar

Sreeram Viswanath Balakrishnan

Raghavendra Singh

Tanveer Afzal Faruquie

Hima Prasad Karanam

Vikas Raykar

Rose Catherine Kanjirathinkal

Manjula Golla Hosurmath

Amit K Singh

Krishna Prasad Chitrapura

Ayushi Dalmia

Ankur Gandhe

Gaurav Pandey

Dhiraj Madan

Saket Saurabh

Manish Anand Bhide

Sriram Kolinjivadi Padmanabhan

Amit Anil Nanavati

Rahul Gupta

Samuel Jonathan Thomas

Shourya Roy

Karthik Visweswariah

George Andrei Saon

Hong-Kwang Jeff Kuo

Brian E D Kingsbury

Robert Leslie Yates

Rema Ananthanarayanan

Priyanka Agrawal

Joy Mustafi

Muhtar B Akbulut

Girish Venkatachaliah

Shajith I Mohamed

Thomas A Hampp-Bahnmueller

Dan O'Connor

Calvin Stacy Powers

Kshitij P Fadnis

Scott Holmes

Hui Wan

Marvin Mendelssohn

Song Feng

Zoltan Tueske

Li Zhu

Kiran Mehta

Ross Warren Judd

Jayadeva

Ganesh Ramakrishnan

Angel Marie Smith

Andy James Stoneberg

Jaikrishnan Hari

Nikhil Verma

Sandeep Singh

Shubham Toshniwal

Karthik Vesweswariah

Nimit Kumar

Himanshu Pant

Jeffrey M Achtermann

Shantanu R Godbole

Vineet Kumar

In an approach to improve abstract summarization with question and answer rewards embodiments generate, by a question and answer generator, questions and answers corresponding to a generated summary. Further, embodiments evaluate received answers for the generated questions by analyzing received answers associated with the generated summary against answers received for an original summary, and calculate a reward based on the similarity between answers associated with generated summary and the original summary. Additionally, embodiments update the generation model by applying the calculated reward to further train the summary generation model.

Enhancing natural language processing accuracy in computer systems

An example system includes a processor to receive a summary of a conversation to be generated. The processor can input the summary into a trained summary-grounded conversation generator. The processor can receive a generated conversation from the trained summary-grounded conversation generator.

Conversation generation using summary-grounded conversation generators

A method, computer system, and a computer program for conducting a conversational search. In one embodiment, the method includes monitoring a dialogue involving at least one user and capturing user utterances provided during the dialogue. These user utterances are then analyzed and classified according to the context of the dialogue. The dialogue is intervened upon the determination that a user needs additional information and/or upon execution of an action on behalf of the user and based on the plurality of user utterances and context. The Required information may be provided back to the user using Documentation Recommendation Module. The Documentation Recommendation Module determines a valid resource recommendation as determined by a combination of the context and a resource that includes additional information.

Techniques for predicting a personalized URL document to assist a conversation

Methods, systems, and computer program products for context-based response generation are provided herein. A method includes: obtaining conversation logs comprising agent responses matched to contexts and a set of designated responses that are not matched to the contexts; replacing at least a portion of the agent responses with the designated responses to form modified conversation logs; training a first model, using the modified conversation logs, to output a designated response in the set for a given context and a second model, using the historical conversation logs, to output one of the agent responses for a given context; identifying one or more new responses based at least in part on the output of the second machine learning model for a particular one of the contexts; and retraining the first machine learning model based at least in part on the one or more new responses.

Context-based response generation

A system may include a memory and a processor in communication with the memory. The processor may be configured to perform operations. The operations may include generating a content page and selecting an annotation container for the content page. The operations may include configuring the annotation container and inputting content into the annotation container. The operations may include submitting at least one attribute to the annotation container to associate the attribute with the content. The operations may include including the content and the at least one attribute in container data and extracting a model from the container data. The operations may include importing the model into a dialog skill and embedding the dialog skill into a user interface.

Incepting conversational ability in content management systems

Audio signals representing a current utterance in a conversation and a dialog history including at least information associated with past utterances corresponding to the current utterance in the conversation can be received. The dialog history can be encoded into an embedding. A spoken language understanding neural network model can be trained to perform a spoken language understanding task based on input features including at least speech features associated with the received audio signals and the embedding. An encoder can also be trained to encode a given dialog history into an embedding. The spoken language understanding task can include predicting a dialog action of an utterance. The spoken language understanding task can include predicting a dialog intent or overall topic of the conversation.

Integrating dialog history into end-to-end spoken language understanding systems

A processor may transcribe an electronic representation of a dialogue. The processor may identify one or more question-answer pairs from the electronic representation. The processor may generate based upon the one or more identified question-answer pairs a summary of the dialogue. The processor may display the summary of the dialogue to a user.

Summary generation guided by pre-defined queries

Embodiments herein describe an automated process for identifying tasks that can be converted from primarily being driven by a human agent into bot-driven RPAs. This disclosure describes techniques for automatically generating RPAs by monitoring the actions of the agent and identifying corresponding events. These events can be performed on unstructured data sources and structured data sources. The events are clustered into different tasks where pattern detection techniques are used to identify common (or overlapping) events and causal relationships between the events. Using these causal relationships, the system can map events performed on unstructured data sources to events performed on structured data source. The system can then generate respective bot-RPAs for the tasks. The system can monitor, in real-time, data received from unstructured and/or structured data sources and when it identifies data for one of the bot-RPAs, automatically execute the RPA to perform the task.

Unstructured extensions to rpa

According to one embodiment, a method, computer system, and computer program product for solution guided generation of responses for dialog systems is provided. The embodiment may include receiving, by a processor, first voice data associated with a first user utterance in conversation in a guided dialog system. The embodiment may include identifying from the first voice data a first topic of a set of topics associated with the first user utterance. The embodiment may include identifying a first solution associated with the first topic. The first solution having one or more solution segments for accomplishing a task related to the topic. The embodiment may include generating a first response for a second user based on a first solution segment of the first solution and the first voice data.

Solution guided response generation for dialog systems

Embodiments of the invention are directed to a computer-implemented method of responding to an inquiry received electronically at a synchronous conferencing system (SCS). A non-limiting example of the computer-implemented method includes, based at least in part on the inquiry, using a processor of the SCS to generate a script having one or more script computer instructions. The processor is used to execute the script computer instructions to generate script computer instruction results, wherein the script computer instruction results include inquiry response information that is responsive to the inquiry. Based at least in part on the inquiry and the inquiry response information, the processor is used to generate an inquiry response that is responsive to the inquiry.

Cognitively learning to generate scripts that simulate live-agent actions and responses in synchronous conferencing

In an approach for a cooperative build and content annotation system for conversational design of virtual assistants, a processor formulates a build context based on a build activity of a user. A processor formulates one or more content queries based on the build context. A processor builds a content index by augmenting a text-search index with a neural Information Retrieval (IR) index. A processor searches the content index using the one or more content queries to identify content relevant to the build context. A processor determines at least one recommendation for the user based on heuristic rules applied to the build context and the identified content, wherein each recommendation is a build suggestion or a content annotation suggestion.

Cooperative build and content annotation for conversational design of virtual assistants

A method and system. Target clusterability is calculated as an average of a respective clusterability of at least one target data item comprised by a target domain. Target-side matchability is calculated as an average of a respective matchability of each target centroid of the target domain to source centroids of a source domain, wherein the source domain comprises at least one source data item. Source-side matchability is calculated as an average of a respective matchability of each source centroid of said source centroids to the target centroids. Source-target pair matchability is calculated as an average of the target-side matchability and the source-side matchability. Cross-domain clusterability between the target domain and the source domain is calculated as a linear combination of the calculated target clusterability and the calculated source-target pair matchability. The cross-domain clusterability is transferred to a device.

Cross-domain clusterability evaluation for cross-guided data clustering based on alignment between data domains

Techniques facilitating interpretation of deep neural model based dialogue agents are provided. In one example, a computer-implemented method comprises extracting, by a device operatively coupled to a processor, features from a dialogue model independently from the dialogue model, the features comprising input features provided to the dialogue model and output features produced via the dialogue model in response to the input features, resulting in extracted features; and analyzing, by the device, a dialogue context associated with the extracted features by identifying pairwise interactions between respective ones of the extracted features.

Interpretation of deep neural model based dialogue agents

In an approach to creation and execution of dialog shortcuts, responsive to detecting initiation of a dialog, an utterance is received from a user. Whether the utterance contains an objective of the user is determined, where the objective is chosen from a group including create a shortcut, execute the shortcut, modify the shortcut, and delete the shortcut. Responsive to determining that the utterance contains the objective, the objective is implemented.

Dialog shortcuts for interactive agents

Methods, systems, and computer program products for generating dialog system workspaces are provided herein. A computer-implemented method includes obtaining (i) a set of policy documents and (ii) a set of initial questions; identifying at least one of the policy documents in the set of policy documents that is relevant to answering a given one of the initial questions in the set of initial questions; generating, based at least in part on an analysis of said identified policy document, (i) at least one follow-up question to said given initial question and (ii) two or more candidate answers to said at least one follow-up question; generating a dialog tree comprising at least (i) a parent node corresponding to the at least one follow-up question and (ii) child nodes corresponding to the two or more candidate answers; translating the dialog tree into a dialog workspace; and deploying the dialog workspace in an intelligent dialog system.

Generating dialog system workspaces

Growing community of inventors

New Delhi, India

Sachindra Joshi