Visual patent trend analysis for informed decision making in technology management

Abstract

Patent information is increasingly important for decision makers. Their demand for exploratory trend and competitor analysis poses new challenges with respect to the processing and visualization of patent data. We present PatStream: a highly interactive approach for decision support through patent exploration, which offers a streamgraph-based visualization for trends at different levels of abstraction and facilitates the combined analysis of their various aspects, including patent applicants, IPC distributions and innovativeness. PatStream integrates powerful natural language processing techniques for concept extraction and patent similarity assessment to allow for content-oriented visualization and analysis.