| Summary: | 碩士 === 國立高雄第一科技大學 === 企業管理研究所 === 99 === Patent’s quality verification could be accessed via the process of litigation. Compared to other industries, computer industrial technologies are prone to patent infringement litigation. The small and medium enterprises (SMEs), which possess the essential patents, are commonly found filing the patent litigation towards the large companies. Therefore, it is important to improve the patent quality, not only from the technology itself, but also through a good patent writing for providing itself a comprehensive protection and higher patent quality. In this study, the computer industry''s U.S. patent infringement lawsuits, patent quality of impact-related indicators; another collection of non-infringement of U.S. patent litigation related to patents, as control sample data. We would adopt the patents to analyze the correlation between the indicators and conduct the binary logistic regression analysis on the litigation and non-litigation cases to identify a more appropriate patent quality prediction model.
The results have shown that a total of five indicators (the number of patent, the number of cited prior art abroad, the total number of cited patents, the number of the priority rights, and the time period of the patent maintenance), have a significant positive correlation with the patent quality. A positive correlation is also found between the patent indicators, the prospective of the patent technology and the possibility rate of its future litigation. These results possess its predictability when the established binary logistic regression analysis model verified by the litigated and non-litigated patents of the particular industry.
|
|---|
