How do patents' intrinsic properties associate with litigation?

this paper aims to devise a novel method for identifying how patent characteristics associated with patent litigation. Large-scale patents granted by the USPTO together with litigation data are converted to items which associations are analyzed by association rule mining. The items considered in this research, for example, Assignee Country, Assignee Country Count, Inventor Country, Inventor Country Count, Number of Patent Reference, Number of Forward Citation, Number of NPR, Number of Foreign Reference, and Numbers of Claim. Obtained with the association relationship depicted by the three variables, 1) Confidence, 2) Support, 3) Lift, an associated network of patents' characteristics can be visually created to understand the landscape of associative relations. The association results uncovered in this research provide a channel to assess whether or not a patent is likely to involve in litigation. Therefore, the influence of patent's intrinsic property hidden in patent's characteristics on the likelihood of infringement can be analyzed.