Notice: Undefined index: linkPowrot in C:\wwwroot\wwwroot\publikacje\publikacje.php on line 1275
[119800] Artykuł: The analysis of Polish patent applications in the solar energy technology with the use of text mining methodologyCzasopismo: Scientific Papers of Silesian University of Technology – Organization and Management Series Zeszyt: 162, Strony: 9-34ISSN: 1641-3466 Opublikowano: Grudzień 2022 Autorzy / Redaktorzy / Twórcy Grupa MNiSW: Publikacja w czasopismach wymienionych w wykazie ministra MNiSzW (część A) Punkty MNiSW: 70 Pełny tekst Keywords: solar energy patent analysis  patent documents processing  patent clustering  outliers detection  Polish Patent Office  |
Purpose: Knowledge management belongs to the most important elements of organisational management, including manufacturing enterprises. Patent information plays an increasingly important role in this area. Identification of the main directions of invention activity may inspire new product and process ideas, and can help to improve existing solutions. The above is particularly important in the energy sector, which is currently struggling with increasing problems. In this context, solar energy is the subject of interest to inventive communities. The paper discusses patent applications related to solar energy, taking up the task of discovering the main tendencies of technological solutions in this area. Design/methodology/approach: In the work, a pilot study of the research aimed to indicate the directions of technological development in the field in Poland was undertaken. Shortened descriptions of selected patent documents from the Polish Patent Office (PPO) were the subject of the investigation. The descriptions were reduced to the form of a vector space model by using text mining tools. The exploration of such prepared data was done applying unsupervised text mining techniques. Hierarchical cluster analysis enabled the identification of groups of similar inventions. An algorithm to detect outliers within individual patent groups was also developed and applied. Findings: Five patent clusters were identified covering the following thematic areas: PV panel designs, PV panel component designs, the improvement of solar-heat conversion device performance, and solar collector designs. Six patent applications stood out thematically in four of the five clusters. Research limitations/implications: The research is limited to a selected number of patent documents form PPO. However, the presented method and research area are promising. It is planned to extend the analyses to a larger set of patent documents and solve the problem related to the language uniformity of patent applications along with merging data from various sources. In this aspect, a full patent description will be consider as well. Originality/value: In relation to solar energy issues, main patent areas and patent outliers that may be indicators of special interests of inventors were identified. In relation to methodology issues, new solutions within consecutive research steps were proposed.