International Business Machines Corporation

Nicholas Mattei

Achille Belly Fokoue-Nkoutche

Maria Chang

Chulaka Gunasekara

Kartik Talamadupula

Kshitij P Fadnis

Ibrahim Abdelaziz

Pavan Kapanipathi Bangalore

Debarun Bhattacharjya

Bassem Makni

Veronika Thost

Avinash Balakrishnan

Siva Sankalp Patel

One embodiment of the invention provides a method for natural language processing (NLP). The method comprises extracting knowledge outside of text content of a NLP instance by extracting a set of subgraphs from a knowledge graph associated with the text content. The set of subgraphs comprises the knowledge. The method further comprises encoding the knowledge with the text content into a fixed size graph representation by filtering and encoding the set of subgraphs. The method further comprises applying a text embedding algorithm to the text content to generate a fixed size text representation, and classifying the text content based on the fixed size graph representation and the fixed size text representation.

Infusing knowledge into natural language processing tasks using graph structures

Techniques for multi-attribute evaluation of narratives are provided. Inputs are obtained representing: (i) at least one historical dataset of events; (ii) a set of candidate narratives, wherein each candidate narrative is a potential future event sequence; and (iii) a query, wherein the query comprises one or more events of interest to a user. Attribute scores are computed for at least a subset of the candidate narratives based on at least a portion of the obtained input. One of the attribute scores comprises a plausibility attribute score representing a measure estimating the likelihood that a given candidate narrative will occur in the future. Another one of the attribute scores comprises a surprise attribute score representing a measure estimating how surprising a given candidate narrative will be to the user.

Multi-attribute evaluation of narratives using event datasets

Growing community of inventors

White Plains, NY, United States of America

Nicholas Mattei