Linguistic summarization of structured patent data

Thesis Type: Post Graduate

Institution Of The Thesis: Gazi Üniversitesi, Bilişim Enstitüsü, Turkey

Approval Date: 2017


Consultant: DİYAR AKAY


The most important reason behind the patent developments has been the protection of valuable information. Examining patent data in detail, it appears that valuable information contains too many and complex patterns from industrial, technical, economic and scientific point of view. It is very important to extract hidden patterns from large-size data such as patent databases, to obtain data that can be easily understood and used by people. In this thesis study, linguistic summarization, one of the descriptive data mining techniques, was used to reveal the relationships between structured patent data. For this purpose, the patent data is obtained by the European Patent Office via the online database PATSTAT. The results obtained by fuzzy linguistic summarization for this thesis study are compared with the hypothesis of patent data intensively discerned in the literature. The correlations that are claimed to be existed between the patent data in the comparison result show a great deal of parallelism with the relations obtained using the linguistic summarization method.