Development of a Patent Matching System Using a Hybrid Approach

DOI

10.2991/jcis.2006.52How to use a DOI?

Keywords

patent matching, data mining, patent analysis

Abstract

There were many researches about applying various data mining or text mining tools to patent analysis, and there were many scholars and experts have verified the accuracy and the feasibility of those tools. However, since mining tools always tried to analyze the content using some mathematic methodology, such as linguistic algorithms, they neglect the fact that patent records are combinations of both structured and non-structured data; it contains not only the non-structured descriptive text but also many structured data related to each patent, such as inventors, assignees and citation information… etc. In another word, mining methodology tent to neglect this import features of patent records and handled them as pure text. This paper proposes a hybrid approach to conduct patent matching process. In this study, an experimental prototype call PMS (Patent Matching System) was developed by composing both data matching and mining approach. By entering several origin patents, the PMS will scan the patent database to generate a similarity ranking table, and then patents that most similar to those origin patents will be suggested to the user. As our sample testing reveals, the PMS achieved a remarkable patent matching capability, and show potential for further improvement.

Copyright

© 2006, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - SuHoun Liu
AU  - Hsiu-Li Liao
AU  - Chou-Chih Hsieh
PY  - 2006/10
DA  - 2006/10
TI  - Development of a Patent Matching System Using a Hybrid Approach
BT  - Proceedings of the 9th Joint International Conference on Information Sciences (JCIS-06)
PB  - Atlantis Press
SP  - 215
EP  - 218
SN  - 1951-6851
UR  - https://doi.org/10.2991/jcis.2006.52
DO  - 10.2991/jcis.2006.52
ID  - Liu2006/10
ER  -