원문정보
초록
영어
One of the most important techniques to improve information management on the web in order to obtain better understanding of user's behaviour is clustering web data. Currently, the rough approximation-based clustering technique has been used to group web transactions into clusters. It is based on the similarity of upper approximations of transactions to merge between two or more clusters. However, in reviewing the technique, it has a weakness in terms of processing time in obtaining web clusters. In this paper, an alternative technique for grouping web transactions using rough set theory, named RMF is proposed. It is based on the rough membership function of a transaction similarity class with respect to the other classes. The two UCI benchmarks datasets are opted in the experimental processes. The experimental results reveal that the proposed technique has an benefit of low time complexity as compared to the baseline technique up to 67 %.
목차
1. Introduction
2. Rough Set Theory
2.1. Information System
2.2. Indiscernibility relation
2.3. Set Approximations
2.4. Rough membership function
3. Analysis of Data Clustering Technique Proposed by (De & Krishna, 2004)
4. The Proposed Technique
4.1. The computational complexity
4.2. Example
5. Results and Discussion
6. Conclusion
References