The Study of Applying Data Mining to Categorize Books

• Main Authors: Ke-Jiang Li, 李克強
• Other Authors: Huei-Chung Chu
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/14618755203218472675

碩士 === 華梵大學 === 資訊管理學系碩士班 === 94 === 　　Book categorization has been a man-made work in library for decades. Since the data mining technology has been advanced and applied to many domains widely, the library researchers also began to leverage this method in offering readers better services, analyzing the reader’s behavior, and cataloguing the books. However, this issue didn’t acquire much attention to study and drill down. In this paper, we applied the data mining technology to book categorization, expect to improve the procedures of book categorization, and reduce the work load of librarian as well. 　　This research started from the survey of operational mode in book classifier system, and then create a two-phase analysis model. In the first phase, the Apriori algorithm was applied to count the probability and gain the class with the higher probability, then use the Naïve Bayes classification algorithm to derive the most probability class in the group. Finally, the model was verified by the library of Huafan University, result show that from the assistant viewpoint, the technology of data mining can help the librarian to reduce their working load.