The Large Aperture Optical Elements patent search system based on Domain Knowledge Organization System

Effective use of patent information provides a multiplier effect in product design and new technology development. This paper reports research on using an open-source extraction tool—General architecture for text engineering (GATE) and word split software—from the Institute of Computing Technology a...

Full description

Saved in:
Bibliographic Details
Published inWorld patent information Vol. 35; no. 3; pp. 209 - 213
Main Authors Tian, Liang, Zhiping, Yang, Zhengyin, Hu
Format Journal Article
LanguageEnglish
Published Elsevier Ltd 01.09.2013
Subjects
Online AccessGet full text
ISSN0172-2190
1874-690X
DOI10.1016/j.wpi.2013.04.007

Cover

More Information
Summary:Effective use of patent information provides a multiplier effect in product design and new technology development. This paper reports research on using an open-source extraction tool—General architecture for text engineering (GATE) and word split software—from the Institute of Computing Technology and Chinese lexical analysis system (ICTCLAS) to assist the expert in acquiring and marking of feature word groups from the abstracts and claims of the patents on Large Aperture Optical Elements. Then, it used the experts feature word groups, which were formalized by INSPEC control words, to construct a Domain Knowledge Organization System (DKOS). Base on the DKOS, a retrieval module of patent information is constructed, which has practical significance for designers to design products and develop new technology. The system is a visual application system, for example, which can filter patent documents by topics and retrieve relevant topics via sample text. [Display omitted] •Improved patent database creation and retrieval system demonstrated.•Marking of features and control word groups to form the database.•Semantic search function used for efficient retrieval.•Applied to the field of optical instrument technology.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0172-2190
1874-690X
DOI:10.1016/j.wpi.2013.04.007