International Business Machines Corporation

Benjamin Sznajder

Sachindra Joshi

Guy Feigenblat

Chulaka Gunasekara

David Carmel

David Konopnicki

Noam Slonim

Ranit Aharonov

Doron Cohen

Artem Spector

Jatin Ganhotra

Dafna Sheinwald

Ilya Shnayderman

Ella Rabinovich

Jonathan Mamou

Sivan Yogev

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 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

Method and apparatus for training and using a classifier for words. Embodiments include receiving a first plurality of sentences comprising a first word that is associated with a class and a second plurality of sentences comprising a second word that is not associated with the class. Embodiments include training a classifier using positive training data for the class that is based on the first plurality of sentences and negative training data for the class that is based on the second plurality of sentences. Embodiments include determining a measure of correlation between a third word and the class by using a sentence comprising the third word as an input to the classifier. Embodiments include using the measure of correlation to perform an action selected from the following list: selecting content to provide to a user; determining an automatic chat response; or filtering a set of content.

Classification engine for learning properties of words and multi-word expressions

A method and system for searching numerical terms are provided. The method includes processing a numerical term to a stem, the stem being in the form of a number, and a distance measurement of the numerical term to the stem, wherein a numerical term is a string of characters identified as a number by a numeric parser, and indexing the numerical term in a search index by the stem for search retrieval. The method includes receiving a query and processing the query to divide it into terms, including a numerical term. The numerical terms of the query are also processed to a stem and a distance measurement and document results are retrieved for the stem from an index.

Method and system for searching numerical terms

A method for information retrieval includes extracting from a video document visual data items and textual data items that occur in the document at respective occurrence times. Indexing records, which index both the visual and the textual data items by their respective occurrence times, are constructed and stored in a memory.

Unified inverted index for video passage retrieval

A computer-implemented method for searching a corpus of documents includes defining a query as a twig including a root annotation operator having an associated tag specifying a span and having an associated expression indicative of one or more terms whose occurrence within the span will satisfy the query. An object is recursively selected from a group of objects that consists of the tag and the expression, and is used in advancing through the corpus until a candidate document is found that contains the tag and satisfies the expression. The candidate document is evaluated to determine whether the one or more terms indicated by the expression occur within the span in the candidate document so as to satisfy the annotation operator.

Growing community of inventors

Jerusalem, Israel

Benjamin Sznajder