Clustering documents on case vectors represented by predicate-argument structures - applied for eliciting technological problems from patents

Patent analysis is useful to understand the trends of technological problems and develop strategies for technologies. Here patent classification is a method to support the analysis. The purpose of this study is to propose a method for patent classification, with the use of hierarchical clustering based on the structural similarity of problems to be solved. The structural similarity can be calculated with case vectors based on predicate-argument structures of the contents of the patents. The interview survey indicated that this classification plays an essential role in analogical problem solving, by allowing visualization of similar technological problems.