Patents by Inventor Gareth A. Latty

Gareth A. Latty has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Schema generation using natural language processing
    Patent number: 10268699
    Abstract: In a method for generating a schema for a corpus of data, a first corpus of data is received, wherein the first corpus of data includes unstructured text. A processor identifies a set of one or more entity relationships within the first corpus of data, wherein an entity relationship comprises a first entity, a second entity, and a specified relationship between the entities. A processor compares the set of one or more entity relationships to a second corpus of data, wherein the second corpus of data includes text of a subject matter different than the corpus of data. A processor determines a score for each entity relationship based on the comparison to the second corpus of data. A processor generates a schema for the first corpus of data based on the score for each entity relationship of the set of one or more entity relationships.
    Type: Grant
    Filed: March 10, 2016
    Date of Patent: April 23, 2019
    Assignee: International Business Machines Corporation
    Inventors: Matthew C. Farenden, Gareth A. Latty
  • SCHEMA GENERATION USING NATURAL LANGUAGE PROCESSING
    Publication number: 20160283523
    Abstract: In a method for generating a schema for a corpus of data, a first corpus of data is received, wherein the first corpus of data includes unstructured text. A processor identifies a set of one or more entity relationships within the first corpus of data, wherein an entity relationship comprises a first entity, a second entity, and a specified relationship between the entities. A processor compares the set of one or more entity relationships to a second corpus of data, wherein the second corpus of data includes text of a subject matter different than the corpus of data. A processor determines a score for each entity relationship based on the comparison to the second corpus of data. A processor generates a schema for the first corpus of data based on the score for each entity relationship of the set of one or more entity relationships.
    Type: Application
    Filed: March 24, 2015
    Publication date: September 29, 2016
    Inventors: Matthew C. Farenden, Gareth A. Latty
  • SCHEMA GENERATION USING NATURAL LANGUAGE PROCESSING
    Publication number: 20160283525
    Abstract: In a method for generating a schema for a corpus of data, a first corpus of data is received, wherein the first corpus of data includes unstructured text. A processor identifies a set of one or more entity relationships within the first corpus of data, wherein an entity relationship comprises a first entity, a second entity, and a specified relationship between the entities. A processor compares the set of one or more entity relationships to a second corpus of data, wherein the second corpus of data includes text of a subject matter different than the corpus of data. A processor determines a score for each entity relationship based on the comparison to the second corpus of data. A processor generates a schema for the first corpus of data based on the score for each entity relationship of the set of one or more entity relationships.
    Type: Application
    Filed: March 10, 2016
    Publication date: September 29, 2016
    Inventors: Matthew C. Farenden, Gareth A. Latty