Patent Informatics for Patent Thicket Detection

点击翻译按钮获取中文摘要

ABSTRACTClarkson, GavinNSF-0425116SGER: Patent Informatics for Patent Thicket DetectionPatent thickets are dense webs of overlapping intellectual property rights. In order to commercialize new technology, an organization must understand the components of the thicket. Cumulative innovation in an industry can lead to blocking patents that effectively stifle innovation. Patent pools, an organizational structure where firms can aggregate patent rights into a package for licensing, can work in these circumstances, but may face antitrust challenges. New antitrust guidelines now permit these patent pools in certain circumstances where they are seen as pro-competitive, but there is no methodology for objectively identifying and measuring thickets. Patent pools are hypothesized to arise when patent thickets already exist. Analyzing pools might provide insight into the identification and measurement of thickets. The research utilizes network analysis and comparison of large networks in order to provide insight into circumstances in which thickets can be the basis for pro-competitive patent pools. The work is exploratory in that the patent networks are very large. The preliminary effort has involved modifying standard network density equations to correct for temporal constraints in patents, and a measure of patent network density has been developed. It will be validated and expanded. The work will contribute to the study of large networks and the resulting tool should be useful in the process of technology search. A number of patent pools in technology will provide the data for analysis and testing of candidate techniques for network analysis. The targets include MPEG-4, MP3, DVD Player and DVD Disc, as well as others.