Using synthetic events to identify complex relation lookups
AssigneeInternational Business Machines Corporation
InventorsTimothy A. Bishop, Stephen A. Boxwell, Benjamin L. Brumfield, Nirav P. Desai, Stanley J. Vernier
Life of this patent
Estimating the term…
Abstract
An approach is provided in which a knowledge manager analyzes a corpus of documents based upon relations corresponding to entities in a structured resource and constructs a natural language context associated with a set of the relations. The knowledge manager maps the natural language context to a database query and, in turn, invokes the database query when the knowledge manager matches a question to the natural language context.
Claims9
An information handling system comprising:
- one or more processors;
- a memory coupled to at least one of the processors;
- and
- a set of computer program instructions stored in the memory and executed by at least one of the processors in order to perform actions of:
- creating a structured resource from a corpus of documents, wherein the structured resource comprises a plurality of first member entities and a plurality of second member entities;
- identifying a plurality of sentences in the corpus of documents that each comprises one of the plurality of first member entities and one of the plurality of second member entities;
- constructing a natural language context, for each of the plurality of sentences, based on a sentence context of the first member entity included in the sentence relative to the second member entity included in the sentence, wherein the constructing produces a plurality of natural language contexts that are different from each other;
- generating a plurality of database queries based on the plurality of natural language contexts, wherein each of the plurality of database queries are different from each other;
- creating a plurality of pattern maps that each comprise one of the plurality of natural language contexts and one of the plurality of database queries corresponding to the natural language context;
- in response to matching a question to each of the plurality of natural language contexts, assigning a priority score to each of the plurality of pattern maps based upon a relative amount at which their corresponding one of the plurality of natural language contexts have been matched against one or more previous questions;
- and
- invoking each of the plurality of database queries corresponding to the each of the plurality of the patterns maps in an order based on the corresponding priority score assigned to each of the plurality of the pattern maps, until a data resource match is reached.
A computer program product stored in a computer readable storage medium, comprising computer program code that, when executed by an information handling system, causes the information handling system to perform actions comprising:
- creating a structured resource from a corpus of documents, wherein the structured resource comprises a plurality of first member entities and a plurality of second member entities;
- identifying a plurality of sentences in the corpus of documents that each comprises one of the plurality of first member entities and one of the plurality of second member entities;
- constructing a natural language context, for each of the plurality of sentences, based on a sentence context of the first member entity included in the sentence relative to the second member entity included in the sentence, wherein the constructing produces a plurality of natural language contexts that are different from each other;
- generating a plurality of database queries based on the plurality of natural language contexts, wherein each of the plurality of database queries are different from each other;
- creating a plurality of pattern maps that each comprise one of the plurality of natural language contexts and one of the plurality of database queries corresponding to the natural language context;
- in response to matching a question to each of the plurality of natural language contexts, assigning a priority score to each of the plurality of pattern maps based upon a relative amount at which their corresponding one of the plurality of natural language contexts have been matched against one or more previous questions;
- and
- invoking each of the plurality of database queries corresponding to the each of the plurality of the patterns maps in an order based on the corresponding priority score assigned to each of the plurality of the pattern maps, until a data resource match is reached.
Dependencies are inferred from the claim text.