The software for formation the matrix of technical functions performed by physical effects based on patent database analysis

The article presents a software for forming a matrix of technical functions performed by physical effects based on the analysis of a patent database. For the synthesis of the physical operation principle of new technical systems, physical effects from the knowledge base developed at the CAD Department of VSTU can be used. Physical effects implement technical functions, which in turn constitute the constructive functional structure of the technical system. The automated system is implemented in Python version 3.7.2. TreeTagger is used for morphological analysis, UDPipe is used for syntactic analysis. The correctness of the algorithms was evaluated on a test sample prepared by hand and consisting of 60 patent documents describing 480 technical functions and 20 physical effects. The results obtained: the method for extracting TF showed an accuracy of 0.87 on the test sample, completeness - 0.77 and F-measure - 0.82, the search for the description of the FE functions with an accuracy of 0.92.

1. Korobkin D., Shabanov D., Fomenkov S., Golovanchikov A. Construction of a matrix "Physical effects – Technical functions" on the base of patent corpus analysis. Communications in Computer and Information Science. 2019;1084:52-68.

2. Korobkin D.M., Fomenkov S.A., Kolesnikov S.G., Kamaev V.A. Synthesis of the physical principle of operation of engineering systems in the software environment CPN Tools. Research Journal of Applied Sciences. 2014; 9(11):749-752.

3. Korobkin D.M., Fomenkov S.A., Kolesnikov S.G. Avtomatizaciya processa formirovaniya informacionnogo obespecheniya bazy dannyh fizicheskih effektov. Vestnik komp'yuternyh i informacionnyh tekhnologij. 2005; 3(9):22-25.

4. Korobkin D.M., Fomenkov S.A., Kolesnikov S.G., Voronin Y.F. System of physical effects extraction from natural language text in the Internet. World Applied Sciences Journal. 2013;24(24):55-61.

5. Korobkin D.M., Fomenkov S.A., Kolesnikov S.G. Metod verifikacii sintezirovannoj funkcional'noj struktury posredstvom postroeniya fizicheskogo principa dejstviya tekhnicheskoj sistemy. Modelirovanie, optimizaciya i informacionnye tekhnologii. 2019;7(2-25):97-109

6. Korobkin D.M., Fomenkov S.A., Kolesnikov S.A. Metod sinteza funkcional'noj struktury novyh tekhnicheskih reshenij na osnove dannyh patentnyh massivov. Modelirovanie, optimizaciya i informacionnye tekhnologii. 2019;7(2-25):135-148.

7. Shabanov D.V., Korobkin D.M., Fomenkov S.A., Kolesnikov S.G. Formirovanie matricy "Fizicheskie effekty - Tekhnicheskie funkcii" na osnove dannyh analiza patentnyh massivov. Matematicheskie metody v tekhnike i tekhnologiyah - MMTT. 2019;7:94-99.

8. Uther, William & Mladenić, Dunja & Ciaramita, Massimiliano & Berendt, Bettina & Kołcz, Aleksander & Grobelnik, etc. TF–IDF. 2010. DOI: 10.1007/978-0-387-30164-8_832.

9. Schmid, & Helmut,. TreeTagger - a language independent part-of-speech tagger. 2009.

10. Straka, Milan & Strakova, Jana & Hajič, Jan. UDPipe at SIGMORPHON 2019: Contextualized Embeddings, Regularization with Morphological Categories, Corpora Merging. 2019.