A function-based patent analysis for support of technical solutions synthesis

The paper describes a method designed to extracting structured information from patent data for support of solving a problem of new technical solution synthesis. Ways of technical function implementations extracted from patents are represented by “object-condition-action” model. An extraneous syntactic parser is used for assigning syntactic category to words from patent claims section. A context-sensitive grammar was developed for achieving compliance of parsed text words to “object-condition-action” model components.