International Business Machines Corporation

Mikhail Sogrin

Daniel John McCloskey

Seamus Ronan McAteer

Ahmed M M R Salem

Seamus R Mac An tSaoir

Aditya Mohan

Akihiro Nakayama

Alexander Troussov

Sergey Yevgenyevich Tsvetkov

John Judge

A method and/or system for query expansion may include: providing a set of training data in a given domain in the form of training question texts and training answer texts, identifying disjoint answer words in the training answer text that do not occur in the associated training question text, generating a graph of question word nodes and answer word nodes generated from the set of training data for the given domain in the form of the training question texts and the training answer texts, wherein edges are provided between a disjoint pair of a question word node for a question word in a training question and an answer word node for a disjoint answer word in an associated training answer, and applying spreading activation through the graph to result in a top n most highly activated nodes that are used as candidate words for expansion of a user query input.

Query expansion using a graph of question and answer vocabulary

A method and system are provided for query expansion. The method may include: providing a graph of question and answer word nodes generated from a set of training data for a given domain in the form of training question and answer texts, wherein the answer word nodes are disjoint words that do not occur in both a training question and an associated training answer and wherein edges are provided between a disjoint pair of a training question word and an associated training disjoint answer word, and applying spreading activation through the graph to result in a top n most highly activated nodes that are used as candidate words for expansion of a user query input.

A method and system are provided for adaptive evaluation of meta-relationships in semantic graphs. The method includes providing a semantic graph based on a knowledge base in which concepts in the form of graph nodes are linked by semantic relationships in the form of graph edges. Metadata are encoded in the edges and nodes of the semantic graph, of weightings for measuring a meta-relationship, wherein the meta-relationship applies to the concepts of the semantic graph and is independent of the semantic relationship defined by the edges of the semantic graph. A graph activation is carried out for an input context relating to one or more concepts of the semantic graph, wherein the weightings are applied to a spreading activation signal through the semantic graph to produce a measure of the meta-relationship for a sub-set of concepts of the semantic graph.

Adaptive evaluation of meta-relationships in semantic graphs

Methods, systems and computer program products are provided. Terms of unsubmitted input data entered by a user during composition of submitted input data for the performance of a task are identified. For an identified unsubmitted term, a relevance score indicating the relevance of the unsubmitted term to user intent is determined. The relevance score of the unsubmitted term is determined using a graphical semantic model based on the submitted input data as a representation of user intent. The identified unsubmitted term is selected, for use in supplementing the submitted input data for the performance of the task, based on the determined relevance score.

Utilizing unsumbitted user input data for improved task performance

Mechanisms are provided for latent ambiguity handling in natural language processing. The method may include: providing an annotated semantic graph based on a knowledge base in which nodes representing semantic concepts are linked by semantic relationships, wherein one or more nodes are annotated to indicate a latent ambiguity of the semantic concept; processing a natural language input by activation of the annotated semantic graph; during processing when encountering an annotated node indicating a latent ambiguity of the semantic concept, applying a pre-defined latent ambiguity process to the node; and replacing any annotated nodes indicating a latent ambiguity of the concept remaining in an output of the processing of the natural language input with an alternative node from the semantic graph.

Latent ambiguity handling in natural language processing

A method and system are provided for query expansion. The method may include: providing a graph of question and answer word nodes generated from a set of training data for a given domain in the form of training question and answer texts, wherein the answer word nodes are disjoint words that do not occur in both a training question and an associated training answer and wherein edges are provided between a disjoint pair of a training question word and an associated training disjoint answer word, including providing weightings for the nodes and edges based on frequency data; and receiving a user query input, activating input nodes in the graph for words in the user query input, and applying spreading activation through the graph using the weightings to result in a top n most highly activated nodes that are used as candidate words for expansion of the user query input.

A mechanism is provided for semantic graph augmentation for domain adaptation. A semantic graph is provided based on a knowledge base in which concepts are linked by semantic relationships. A domain corpus of content in a new target domain is analyzed for adaptation with reference to one or more resources that provide syntactic analysis of the domain content that links portions of the domain corpus to existing concepts and links in the semantic graph. Triples are extracted from the analysis of the domain corpus of any two concepts that share a linking syntactic relationship. The triples are evaluated as potential relationship assertions. A relationship assertion is selected for addition as a new link between existing concepts in the semantic graph.

Semantic graph augmentation for domain adaptation

A processor expands a search expression. The processor determines nodes representing query terms of a search expression. The nodes have associated text for search term expansion, and represent at least one concept in a semantic graph of nodes that represents a domain of semantically related concepts. The processor determines i) a center of focus within the semantic graph for the two or more nodes based, at least in part, on a spreading activation in the graph and ii) a contextual relevance for the two or more nodes with respect to node the center of focus. The processor selects, for a query term, a node based on contextual relevance between that node and the query term and expands the search expression using an associated text of that node.

Context sensitive query expansion

This illustrative embodiments provide a mechanism for correcting a phonetically sourced spelling mistake. The mechanism receives a language text string comprising at least one spelling mistake word and transcribes the at least one spelling mistake word into a phonetic form of the spelling mistake word using a phonetic dictionary. The mechanism locates a correctly spelled phonetic form from a phonetic form dictionary having shortest edit distance between characters of the correctly spelled phonetic form word and the phonetic transcription whereby the phonetic form dictionary comprises correctly spelled words and associated phonetic forms of the correctly spelled words. The mechanism substitutes the correctly spelled word for the spelling mistake word in the text string.

Orthographic error correction using phonetic transcription

Mechanisms are provided for modifying an ontology for use with a natural language processing (NLP) task. A model of the ontology is obtained that comprises nodes representing concepts of the ontology and edges between nodes representing relationships between associated nodes of the ontology. A spreading activation operation is performed on the model of the ontology with the spreading activation operation matching a portion of a textual source to a matching node of the model and identifying related nodes to the matching node through edges of the model associated with the matching node to thereby generate an activation network. The activation network is evaluated with regard to a chosen NLP task to determine a performance metric for the NLP task associated with the nodes of the model. Based on the results, one of the model or a configuration of the activation network may be modified.

Automatic evaluation and improvement of ontologies for natural language processing tasks

Method and apparatus for finding the focus of a document. A semantic network includes a plurality of nodes, each representing a concept, and links, connecting the nodes, representing relations between the concepts is used. The method including: providing a list of terms in an input document which match concepts represented by the semantic network, and a frequency value for each matched term indicating the number of times the term appears in the document; mapping each matched term to a referent node or to a plurality of possible referent nodes in the semantic network, and assigning weights to nodes.

Growing community of inventors

Kildalkey, Ireland

Mikhail Sogrin