Ontology-based neural network for patent knowledge management in design collaboration

Amy J.C. Trappey*, Charles V. Trappey, Tzu An Chiang, Yi Hsuan Huang

*此作品的通信作者

研究成果: Article同行評審

40 引文 斯高帕斯(Scopus)

摘要

In order to stimulate innovation during the collaborative process of new product and production development, especially to avoid duplicating existing techniques or infringing upon others patents and intellectual property rights, the collaborative team of research and development, and patent engineers must accurately identify relevant patent knowledge in a timely manner. This research develops a novel knowledge management approach using ontology-based artificial neural network (ANN) algorithm to automatically classify and search knowledge documents stored in huge online patent corpuses. This research focuses on developing a smart and semantic oriented classification and search from the sources of the most critical and well-structured knowledge publications, i.e. patents, to gain valuable and practical references for the collaborative networks of technology-centric product and production development teams. The research uses the domain ontology schema created using Protégé and derives the semantic concept probabilities of key phrases that frequently occur in domain relevant patent documents. Then, by combining the term frequencies and the concept probabilities of key phrases as the ANN inputs, the method shows significant improvement in classification accuracy. In addition, this research provides an advanced semantic-oriented search algorithm to accurately identify related patent documents in the patent knowledge base. The case demonstration analyses 343 chemical mechanical polishing and 150 radio-frequency identification patents sample sets to verify and measure the performance of the proposed approach. The results are compared with the previous automatic classification methods demonstrating much improved outcomes.

原文English
頁(從 - 到)1992-2005
頁數14
期刊International Journal of Production Research
51
發行號7
DOIs
出版狀態Published - 1 4月 2013

指紋

深入研究「Ontology-based neural network for patent knowledge management in design collaboration」主題。共同形成了獨特的指紋。

引用此