System and method for detecting text similarity over short passages
- Summary
- Lead Inventors: Judith KlavansSystem and method for determining similarity in short text segments are described. Methods of natural language processing are described.The method and system comprise of an interface circuit for receiving text segments for comparison. A main processing section is operatively coupled to the interface circuit and operates under the control of a computer program. The program performs operations to determine common primitive features in the text segments. Common composite features in the text segments are determined. A similarity measure based upon primitive and composite features are calculated and an output indicative of the similarity measure is provided.The system and method provide a fine-grained distinction for similarity measures to properly characterize the similarity of two small text segments.
- Detailed Technology Description
- System and method for determining similarity in short text segments are described. Methods of natural language processing are described.The method and system comprise of an interface circuit for receiving text segments for comparison. A main pr...
- *Abstract
-
None
- *Inquiry
- Calvin Chu Columbia Technology Ventures Tel: (212) 854-8444 Email: TechTransfer@columbia.edu
- *IR
- MS99/05/03
- *Principal Investigator
-
- *Web Links
- Patent number: EP1203309Original research paper: Detecting Text Similarity over Short Passages: Exploring Linguistic Feature Combinations via Machine LearningJudith L. Klavans, Ph.D.
- Country/Region
- USA
For more information, please click Here
