Summary: | 碩士 === 元智大學 === 資訊管理學系 === 104 === In recent years, RFID technology has become a widely used technology, it remote reader via radio waves automatically read the contents of the electronic tag, can conduct tracking, theft, wisdom shelves and so on. It takes enormous benefits for industries, and become the main field for many countries to put a lot of R&D in technology development, and many countries apply patents with new technologies to protect its patent quality. Therefore patent quality analysis is important for current various industries.
We collected the patent data on the RFID industry from the patent database of Thomson Innovation, and using the proposed forecasting model to predict the quality of a new published patent. Therefore, we collected 27938 patent data from 1996 to 2013 and adapted the patent quality indicator “Legal Status” to define the level of quality in this study. Next, we use the contents of International Patent Classification(IPC) to obtain the high quality technologies playing roles as features of a patent. Finally, we developed an automatic patent quality forecasting model based on artificial immune network and according to the results of a series of experiments, the proposed approach is efficient than other compared ones.
|