A Study of Effective Mining Frequent Continuity Algorithm in Long Event Sequences

• Main Authors: Ching-hui Su, 蘇津慧
• Other Authors: None
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/29079248319727902456

碩士 === 國立臺南大學 === 資訊教育研究所碩士班 === 93 === Data mining has become a popular method to locate useful information from large databases. Mining frequent continuity is one important part of data mining. Mining frequent continuity can find not only the order of events but also the exact positions of events. Therefore, it is really useful in trend prediction. To reduce returning too many duplicates, this study proposes the QPROWL. The QPROWL method is based on the anti-monotone property of frequency to discover all frequent continuity. According to the simulation results, the QPROWL uses less memory and running time than PROWL. Furthermore, this study extends the QPROWL method to mine frequent continuity without re-executing QPROWL when the windows are updated.