Comparative study on book similarity measurement method based on characters in the body of a book written in Korean
Hyun-Ho Lee, Min-Ha Hong, Kee-Won Kim, Seung-Hoon Kim · 2017
This paper presents a comparative study on similarity measurement method based on characters in the body of a book written in Korean. We deduce using ratio of characters in the body of a book for measuring similarity and compare results of book similarity using three different similarity measurement methods through the experiment. And we measure similarity using SMC (Simple Matching Coefficient), Jaccard coefficient and cosine similarity based on extracted characters in six books which are history category. The proposed method can measure book similarity more detail than existing research and result of measurement using cosine similarity is more accurate than other two methods because cosine similarity considers not only presence of characters also frequency of characters in the body of a book. If measuring book similarity based on characters in the body of books, we expect quick and detail performance at books which are categorized history, biography, etc.