Thesis Type: Doctorate
Institution Of The Thesis: Gazi Üniversitesi, Fen Bilimleri Enstitüsü, Turkey
Approval Date: 2016
Student: SERKAN GENÇ
Consultant: DİYAR AKAYAbstract:
Although a plenty of techniques such as link prediction, clustering, position analysis have been proposed to analyze social networks, few studies have been addressed to transform social networks data into the knowledge in the form of linguistic summaries. Linguistic summarization, capable of extracting explicit and concise summary from a collection of raw data, is more compatible with human cognitive mechanism. In this thesis, the application of linguistic summarization to information networks which is one of the descriptive data mining/knowledge discovery techniques, linguistic summary forms for information networks, and methods for evaluating the degree of truth of suggested linguistic summary forms are aimed. In order to do these, for the first time in the literature, iteration, resumption, reciprocal and branching processes based linguistic summary forms taking into account both the attributes of the nodes and the relations between them are proposed. Then, methods for evaluating the degree of truth of linguistic summary forms suggested by taking advantages of generalized quantifiers, especially semi-fuzzy quantifiers and polyadic quantifiers, are developed. Finally, the suggested methods are used for the linguistic summarization of the international trade network.