Association Rule Visualization by Structured Association Map
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Association rule mining is one of the most popular data mining techniques, and its aim is to extract the association rules, the cause-and-effect relations between the items, from the given transaction data. Several algorithms such as apriori and its variants have been developed in order to extract the association rules in efficient way, however, they often produce the plethora of the association rules that is difficult for the analyzers to interpret and exploit. To address this issue, this paper aims to propose a visualization method called structured association map. The structured association map is a variant of the well known cluster hear map, and it focuses on ordering the items in more meaningful way. The structured association map and the cluster heat map have in common that the dendrogram obtained by hierarchical clustering is appended to the matrix for data visualization and the items are ordered according to the dendrogram. On the contrary, the primary difference between the two visualization methods lies in the way the hierarchy of the column items is generated. In structured association map, the row items are ordered at first and their order is considered in ordering the column items, while the row items and column items of the cluster heat map are ordered in similar manner. Consequently, the structured association map can represent both antecedents and consequents of the association rules in more effective way, and it is expected to help the analyzers to understand the structure of the extracted association rules more conveniently.
Abstract
Association rule mining is one of the most popular data mining techniques, and its aim is to extract the association rules, the cause-and-effect relations between the items, from the given transaction data. Several algorithms such as apriori and its variants have been developed in order to extract the association rules in efficient way, however, they often produce the plethora of the association rules that is difficult for the analyzers to interpret and exploit. To address this issue, this paper aims to propose a visualization method called structured association map. The structured association map is a variant of the well known cluster hear map, and it focuses on ordering the items in more meaningful way. The structured association map and the cluster heat map have in common that the dendrogram obtained by hierarchical clustering is appended to the matrix for data visualization and the items are ordered according to the dendrogram. On the contrary, the primary difference between the two visualization methods lies in the way the hierarchy of the column items is generated. In structured association map, the row items are ordered at first and their order is considered in ordering the column items, while the row items and column items of the cluster heat map are ordered in similar manner. Consequently, the structured association map can represent both antecedents and consequents of the association rules in more effective way, and it is expected to help the analyzers to understand the structure of the extracted association rules more conveniently.
참고문헌 (30)
- R. Agrawal, "Mining associations between sets of items in massive databases", Proceedings of the ACM-SIGMOD 1993 International Conference on Management of Data, pp.207~216 (1993)
- R. Agrawal, "Fast algorithms for mining association rules", Proceedings of the International Conference on Very Large Databases, pp.125~131 (1994)
- P. -N. Tan, "Introduction to data mining" (2005), Addison-Wesley
- B. Nath, "Incremental association rule mining: a survey", Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 3(3), pp.157~169 (2013)
- M. C. Chen, "Mining changes in customer behavior in retail marketing", Expert Systems with Applications, 28(4), pp.773~781 (2005)
- H. Kim, "A framework for tag-aware recommender systems", Expert Systems with Applications, 41(8), pp.4000~4009 (2014)
- J. Jeon, "Product failure pattern analysis from warranty data using association rule and Weibull regression analysis: a case study", Reliability Engineering and System Safety, 133, pp.176~183 (2015)
- B. Liu, "Pruning and summarizing the discovered associations", Proceedings of the fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.125~134 (1999)
- Y. A. Sekhavat, "Visualizing association rules using liked matrix, graph and detail views", International Journal of Intelligence Science, 3, pp.34~49 (2013)
- Y. Djenouri, "Pruning irrelevant association rules using knowledge mining", International Journal of Business Intelligence and Data Mining, 9(2), pp.112~144 (2014)
- B. Lent, "Clustering association rules", Proceedings of 13th International Conference on Data Engineering, pp.220~231 (1997)
- 김준우, "유용한 연관 규칙 추출을 위한 시각적 탐색 기반 접근법", 한국컴퓨터정보학회논문지, 18(9), pp.177~187 (2013), 한국컴퓨터정보학회
- D. A. Keim, "Information visualization and visual data mining", IEEE Transactions on Visualization and Computer Graphics, 7(1), pp.100~107 (2002)
- M. C. F. De Oliveira, "From visual data exploration to visual data mining: a survey", IEEE Transactions on Visualization and Computer Graphics, 9(3), pp.378~394 (2003)
- P. C. Wong, "Visualizing association rules for text mining", Proceedings of the 1999 IEEE Symposium on Information Visualization, pp.120~123 (1999)
- C. Romero, "RM-Tool: a framework for discovering and evaluating association rules", Advances in Engineering Software, 42(8), pp.566~576 (2011)
- R. J. Bayardo, "Mining the most interesting rules", Proceedings of the 5th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.145~154 (1999)
- M. Hahsler, "Visualizaing association rules: introduction to the R-extension package arulesViz, R project module" (2011)
- K. Techapichetvanich, "Advanced Data Mining and Applications", pp.88~95 (2005), Springer Berlin
- P. Buono, "From Integrated Publication and Information Systems to Information and Knowledge Environments", pp.221~223 (2005), Springer Berlin
- L. Yang, "Pruning and visualizing generalized association rules in parallel coordinates", IEEE Transactions on Knowledge and Data Engineering, 17(1), pp.60~70 (2005)
- L. Yang, "Visual exploration of frequent itemsets and association rules", Lecture Notes in Computer Science, 4404, pp.60~75 (2008)
- S. Bornelöv, "Ciruvis: a web-based tool for rule networks and interaction detection using rule-based classifiers", BMC Bioinformatics, 15(1), pp.139~150 (2014)
- M. Hahsler, "Visualizaing association rules in hierarchical groups", Proceedings of the 42nd Symposium on the Interface Statistical, Machine Learning and Visualization Algorithms (2011)
- I. Liiv, "Seriation and matrix reordering methods: an historical overview", Statistical Analysis and Data Mining: The ASA Data Science Journal, 3(2), pp.70~91 (2010)
- L. Wilkinson, "The history of the cluster heat map", The American Statistician, 63(2), pp.179~184 (2009)
- W. H. E. Day, "Efficient algorithms for agglomerative hierarchical clustering method", Journal of Classification, 1(1), pp.7~24 (1984)
- A. Guenoche, "Efficient algorithms for divisive hierarchical clustering", Journal of Classification, 8(1), pp.5~30 (1991)
- 강현경, "집단 건강검진을 위한 계층군집 기반 개인화 피드백 시스템", 한국지식정보기술학회 논문지, 6(4), pp.103~112 (2011), 한국지식정보기술학회
- A. Strehl, "Distance based clustering of association rules", Proceedings of ANNIE 1999, pp.759~764 (1999)