Data mining jako metoda použitelná v oblasti japonských studií

Štefková, Tereza

In this thesis we address the problem of possible utilization of text mining methods in the field of Japanese studies. We review the fundamental text mining approaches and their practical applications in the first part. Then we elaborate on the topic of preprocessing with special focus on techniques used for Japanese and English texts. In the main part of the thesis we apply text mining methods to three concrete research questions relevant in Japanese studies. The first research topic illustrates the technique of clustering applied to works written by two Japanese proletarian authors to reveal interesting topic patterns in their writings. The second topic makes use of the sentiment analysis with the aim of studying the extent of negative sentiment expressed in both foreign and Japanese newspaper articles that refer to Japanese officials' visits to Yasukuni shrine. Finally, we address methods of automatic summarization and their application to Japanese as well as English sample texts. The results obtained are discussed in detail with a special focus on the assessment of viability of the presented methods in Japanese studies.

host :: přihlásit Digitální repozitář
		Hledej		Nový záznam		Nápověda		O repozitáři