CONSTRUCTION, MANIPULATION, AND COMPARISON OF A MULTI-DIMENSIONAL SEMANTIC SPACE

• US 20080052283A1
• Filed: 10/30/2007
• Published: 02/28/2008
• Est. Priority Date: 02/25/2000
• Status: Active Grant

• Current Assignee: RPX Corporation
• Original Assignee: Novell Incorporated (Open Text Corporation)
• Inventors: Jensen, Delos, Carter, Stephen
• Granted Patent: US 8,131,741 B2
• Time in Patent Office: Days
• US Class Current: 707/4
• CPC Class Codes: G06F 16/243 Natural language query formulation
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

A directed set can be used to establish contexts for linguistic concepts: for example, to aid in answering a question, to refine a query, or even to determine what questions can be answered given certain knowledge. A directed set includes a plurality of elements and chains relating the concepts. One concept is identified as a maximal element. The chains connect the maximal element to each concept in the directed set, and more than one chain can connect the maximal element to any individual concept either directly or through one or more intermediate concepts. A subset of the chains is selected to form a basis for the directed set. Each concept in the directed set is measured to determine how concretely each chain in the basis represents it. These measurements for a single concept form a vector in Euclidean k-space. Distances between these vectors can be used to determine how closely related pairs of concepts are in the directed set.

Claims

1. A method for a user of a computer system to find a context to aid in answering a question, the method comprising:

parsing the question into one or more parsed concepts;

establishing one or more chains in a directed set, wherein each chain is rooted at a maximal element in the directed set and extends to a concept in the directed set;

creating a distance vector for the one or more parsed concepts in the directed set, wherein each distance vector includes as its components the measure of how concretely the concept is represented in each chain; and

measuring a distance between the distance vectors for each pair of parsed concepts; and

using the distances between the one or more parsed concepts to establish a context for the question.

2. A method for using a lexicon to submit a refined query input by a user to a query engine, wherein the refined query the method comprising:

parsing the query into one or more parsed concepts;

measuring distances in the lexicon between the one or more parsed concepts;

using the distances between the one or more parsed concepts to establish a context for the query;

refining the query according to the context for the query; and

submitting the refined query to the query engine.