International Business Machines Corporation

Aleksandr Evgenyevich Petrov

John J Thomas

Wanting Wang

Maxime Allard

Vinay R Dandin

Jason D LaVoie

Kristin A Witherspoon

Avijit Chatterjee

Jeffrey R Pratt

Kavita Patil

Romina M J Jose

Edward T Winchester

Ricardo Balduino

A question-and-answer system directed to a specific domain optimally utilizes reference documents that are semantically complete for that domain. Semantic completeness of a document is assessed using quality control questions (provided by subject matter experts) applied to the Q&A system followed by analysis of the proposed answers. That analysis is carried out using a cogency module having a feedforward neural network which receives metadata features of the document such as document ownership, document priority, and document type. A domain-optimized corpus for the Q&A system is built by so assessing multiple documents in a document collection, and adding each reference document that is reported as being semantically complete to the domain-optimized corpus. Thereafter, the deep learning question-and-answer system can receive a natural language query from a user, find a responsive answer in the documents while applying the domain-optimized corpus, and provide that answer to the user.

Providing semantic completeness assessment with minimal domain-specific data

A method of training a neural network includes receiving a text corpus containing a labeled portion and an unlabeled portion, extracting local n-gram features and a sequence of the local n-gram features from the text corpus, processing the text corpus, using convolutional layers, according to the local n-gram features to determine capsule parameters of capsules configured to preserve the sequence of the local n-gram features, performing a forward-oriented dynamic routing between the capsules using the capsule parameters to extract global characteristics of the text corpus, and processing the text corpus according to the global characteristics using a long short-term memory layer to extract global sequential text dependencies from the text corpus, wherein parameters of the neural network are updated according to the local n-gram features, the capsule parameters, global characteristics, and global sequential text dependencies.

Classification of sparsely labeled text documents while preserving semantics

In a general purpose computer, a method of extracting snippets includes receiving textual content and a plurality of available topics, dividing the textual content into a plurality of snippets, converting each of the snippets to a vector, determining a distance between coadjacent snippets of the plurality of snippets in the textual content, determining an update to the plurality of snippets by merging each of the pairs of coadjacent snippets having a respective distance less than a second threshold, wherein an updated plurality of snippets includes merged snippets, generating a plurality of clusters from the updated plurality of snippets, each cluster associated with one topic selected from the plurality of available topics, and generating, for each of the snippets of the updated plurality of snippets, an affinity score for each of the clusters, each affinity score measuring an assignment strength of a given snippet to a given cluster, and a dominant topic among the at least one identified topic.

Incremental discovery of salient topics during customer interaction

A method, system and computer-usable medium are disclosed for enhancing a classification system to include an automatic classification validation system. The automatic classification validation system takes the classification results and automatically validates them for correctness. More specifically, the automatic classification validation system analyzes the data objects in the categories and, if any outliers are identified, then determines a context of the data object from a plurality of records contained within a certain category to determine the context. The classification validation system then uses the context of the data object to validate the classification.

System and method for automatically validating classified data objects

Growing community of inventors

Acton, MA, United States of America

Aleksandr Evgenyevich Petrov