×

Ranking answers to a conceptual query

  • US 8,898,154 B2
  • Filed: 09/19/2012
  • Issued: 11/25/2014
  • Est. Priority Date: 09/19/2012
  • Status: Expired due to Fees
First Claim
Patent Images

1. A method for ranking answers to a conceptual query, said method comprising:

  • a processor of a computer system receiving the conceptual query, wherein said query comprises a set of queried concepts of a general set of concepts;

    said processor receiving a set of semantic schemas, wherein said set of semantic schemas comprises a received subset of concepts of said set of queried concepts and further comprises a plurality of relationships, wherein a relationship of said plurality of relationships relates two or more concepts of said general set of concepts, wherein a first schema of said set of semantic schemas comprises a first subset of concepts of said received subset of concepts, a second concept of said general set of concepts, a third concept of said general set of concepts, and a first set of relationships of said plurality of relationships, wherein said first subset of concepts comprises a first concept of said received subset of concepts, wherein a first relationship of said first set of relationships relates said first concept to said second concept, wherein said first relationship is associated with a first relationship strength of a plurality of relationship strengths, wherein said first concept is associated with a first concept weight within said conceptual query of a set of concept weights, wherein said second concept is associated with a second concept strength of a set of concept strengths, and wherein said second concept is associated with a second semantic weight in said first schema of a set of semantic weights;

    said processor identifying a first semantic value of said first schema to said query as a function of said first subset of concepts and as a further function of said first set of relationships;

    said processor identifying a first relative relevance of said first schema to said query as a function of said first semantic value of said first schema to said query, whereinsaid identifying a first semantic value of said first schema to said query is a further function of a first set of concept semantic values within said first schema of a plurality of sets of concept semantic values, and wherein said first concept is associated with a first concept semantic value in said first schema of said first set of concept semantic values, and wherein said identifying a first semantic value of said first schema to said query is a further function of a first set of relationship strengths within said first schema of said plurality of relationship strengths, and wherein said first relationship is associated with a first relationship strength of said first set of relationship strengths; and

    whereinsaid first concept semantic value in said first schema is a function of said first relationship strength and is a further function of said second semantic weight in said first schema, wherein said second semantic weight is a function of said second concept strength, wherein said second semantic weight is a further function of a set of second-concept relationship strengths of said plurality of relationship strengths, and wherein a relationship strength of said set of second-concept strengths is associated with a relationship between said second concept and said third concept;

    wherein said identifying a first relative relevance of said first schema comprises the steps of;

    said processor plotting said conceptual query as a first point in a p-dimensional coordinate system and further plotting said first schema as a second point in said p-dimensional coordinate system, wherein an axis of said p-dimensional coordinate system is associated with a concept of said set of queried concepts;

    said processor generating a first p-dimensional tensor from an origin of the p-dimensional coordinate system to said first point;

    said processor generating a second p-dimensional tensor from an origin of the p-dimensional coordinate system to said second point; and

    said processor identifying said first relative relevance of said first schema as a function of a parameter selected from the group comprising;

    a p-dimensional distance between said first point and said second point, a product of said first concept semantic value and said first concept weight within said conceptual query, and a scalar product of said first p-dimensional tensor and said second p-dimensional tensor, wherein said first concept semantic value is associated with said first concept; and

    said processor ranking said first schema among said set of semantic schemas as a function of said first relative relevance.

View all claims
  • 2 Assignments
Timeline View
Assignment View
    ×
    ×