| Summary: | 碩士 === 淡江大學 === 資訊工程學系 === 89 === Nowadays, tremendous amount of data, has been digitized, collected and stored in large and numerous databases. The fast-growing data has far exceeded our human ability for comprehension without powerful tools.. The abundance data, coupled with the need for powerful data analysis tools, has bee described as a date rich but information poor situation. The most important function of a library is to assist the library users or readers in acquiring target books for them. But due to the rapid growth of volumes and books in libraries, it is more difficult to find a book, which matches user’s needs. For this reason, we propose the notion of developing a library recommendatory system to help borrowers finding the book they need.In this paper, we have attained 2 databases from Aletheia University. The first one is the books database, which includes the data of 61160 volumes of books in the university library, and the borrowing history of 14087 students who have borrowed books from the university library. The second one is the students database, which includes records of all students of Aletheia University from the office of administration and registration. From these two databases, we have developed a recommendatory system to recommend related books to the borrowers.Fuzzy theory and clustering methodology are used to group the borrowers into clusters with the similar background. And then we use data mining association rule to analyze the books borrowed by the students of the same cluster, a recommendatory list of books will be produced, so that the borrowers will find references of more books to borrow. Besides, the library recourse may be used more effectively.
|
|---|
