可能性聚類演算法應用於資訊檢索之研究

Information Retrieval System Based on Similarity-Based Clustering Method

張濬杰(JUN-JIE ZHANG)

指導教授 ： 楊敏生

中原大學/理學院/應用數學研究所/碩士(2012年)

https://doi.org/10.6840/cycu201200345

摘要

網際網路的快速普及，時至今日已成為人們生活中的重要工具，由於網路上的資料繁多，在搜尋學術論文時得到的資料量非常龐大，因此一個能幫助資料搜尋者快速找到所需的資訊資料系統，已是一件不可或缺的工作。目前大多數使用聚類演算法建立字詞間關聯規則，皆是使用階層式聚類演算法，因此本研究加入可能性聚類演算法，以期達到更好的分群效果，藉以提供更好的關聯規則。本研究從台灣碩博士論文加值系統中擷取了60篇論文作為樣本，從實驗結果發現，所推論出的關聯規則確實有其良好效果。

關鍵字

階層式聚類演算法 ； 關聯規則 ； 可能性聚類演算法

並列摘要

Because of the rapid spread of Internet, it has become an important tool for human life. Since the range of information on Internet is wide, the amount of information for searching academic papers becomes very large. To create an information system for quickly searching relative knowledge is essential. Most clustering algorithms used for association rules between words are hierarchical clustering. Therefore, in this thesis, we use a robust possibilistic clustering method for achieving better association rules between words. In this study, there are 60 papers retrieved from a website as samples for experimental comparisons. We find that the inferred association rules using our method are indeed with better results.

並列關鍵字

hierarchical clustering algorithm ； association rules ； SCM

參考文獻

[3] G. Salton and C. Buckley, Term-weighting approaches in automatic text retrieval, Information Processing&Management 24 (1988) 513-523.

[4] C.H. Jiang, Building the fuzzy retrieval system based on data mining algorithm, Master thesis, I-Shou University, 2009.

[5] M.S. Yang and K.L. Wu, A similarity-based robust clustering method, IEEE Trans. on Pattern Analysis and Machine Intelligence 26 (2004) 434-438.

[6] J.A. Hartigan, Clustering Algorithms, New York: Wiley, 1975.

[1] ComScore, ComScore Reports Global Search Market Growth of 46 Percent in 2009, http://www.comscore.com/Press_Events/Press_Releases/2010/1/Global_Search_Market_Grows_46_Percent_in_2009.