- 學位論文
以專利本體論工程為基之智財服務防禦式決策支援系統
Using Patent Ontology Engineering to Enable Intellectual Property Defense-based Support System
摘要
二十一世紀是知識經濟的時代，針對公司的無形資產，技術發展與創新是相當重要的知識資本；而技術的擁有者事先一定會對所擁有之技術進行必要的智慧財產權保護，以免遭到侵害後，無法受到法律的保障。專利權是智慧財產權保護的重要一環，它的目的在鼓勵、保護、善用發明與創作，以促進產業發展。從2009-2010年全球競爭力報告顯示，由於瑞士在研發上有相對高的投資且研發的成果被妥善商業化，且每百萬個居民即擁有多達148篇專利，所以瑞士取代了美國成為全球最具競爭力國家。 近年來跨國專利申請與通過件數快速增加，從2004至2006年台灣跨國專利發明數量已佔台灣總專利發明數量的 52.2%，這反映台灣與全球專利研發益趨緊密。如此大量的智慧財產累積，也造就台灣與國外大廠的專利糾紛陸續發生，其所支付的專利授權金額亦是逐年攀升。 無論原告或被告，專利訴訟皆應由被動觀念升級到主動觀念，甚至是預先反應的層次。但在專利侵害攻防方面，除了基本專利搜尋檢索之外，運用先進資訊技術與決策模式進行專利侵害分析與決策支援系統之發展仍相當缺乏。 本研究提出智財服務防禦式決策支援系統，其主軸為以專利本體論工程為基之自動技術群聚與侵害分析。在本體論工程中，主要針對申請專利範圍解析與文件特徵分析，以利領域知識之本體論呈現。在專利技術群聚與侵害分析方法論中，本研究提出專利有效性分析與侵害鑑定前案分析，發展出改善式重疊式分群演算法，利用該演算法進行專利技術群聚及系爭專利之前案與比對專利之檢索認定。亦即，若企業提出反控侵害時，針對所搜尋到的比對專利，設計侵害鑑定比對介面，協助智財人員快速比對全要件原則、均等論與先前技術阻卻等判斷法則。 因此，本研究發展並整合專利本體論與侵害鑑定雛型系統，以資訊系統輔助並引導智財人員系統化解析申請專利範圍，並提供一個以構成要件為基礎的自動侵害比對功能。將專利構成要件與結構關係當作事實，而全要件原則與均等論當作規則，進行事實規則推論。搭配適時之人機介面，提供企業智財服務之防禦式決策支援，協助專利有效性評估與侵害鑑定比對。
並列摘要
In a knowledge based economy, intangible assets (e.g., intellectual properties and innovative technology) are important to the development of enterprise knowledge capital. Patent aims to encourage and protect the use of invention and creation to promote industrial development under global legal rules and orders. Thus, patents are the most critical means of maintaining legal rights of intellectual properties (IPs). The 2009-2010 global competitiveness report indicates that Switzerland became the most competitive nation in the world, because of its relatively high R&D investment and commercialization. Switzerland‘s per million inhabitants own 148 patents, which is the highest in the world, and the number well reflect the R&D capability of a country. In recent decade, Taiwan industry has also focused on the internationalization of R&D with their foreign patents reaching more than 52.2% of total granted patents (well above global average of 15.41%). The increasing patents, applied and approved, attribute to the higher chance of patent litigation and legal disputes. Furthermore, the cost of patent licensing is also growing rapidly over the recent years. Regardless being plaintiffs or defendants, the concept of patent litigation should be upgraded from passive to active and even proactive level. However, other than available strait forward patent search and meta-analysis IT tools, we still lack advanced analytical and decision support methodology and systems for handling patent litigation and protecting IPs. This research will develop patent ontology engineering approach, the improved overlapping patent clustering method and patent infringement analysis. The ontology engineering approach analyzes patent claims structure and extracts patent characteristics for technical domain schema construction. Further, the research will use the developed methods to implement an analytical and decision support prototyping system to analyze related patents and compare their claim elements based on the patent ontology, infringement rules and facts. After the development of the methodology and system, the research will study well known patent litigation cases (e.g., HTC vs. Apple, LGD vs. AUO) and adopt the proposed methods and prototyping system for the case analysis to demonstrate the effectiveness and applicability of the research solutions toward the real legal matter.