데이터 스트림 환경을 위한 유틸리티 기반 웹 방문 패턴의 마이닝 기법 = Efficient Mining of High Utility Web Traversal Patterns Over Data Streams

다국어 초록 (Multilingual Abstract)

Web access sequence mining can discover the frequently accessed web pages pursued by users.
Utility‐based web access sequence mining handles non‐binary occurrences of web pages and extracts more useful knowledge from web logs. However, the existing utility‐based web access sequence mining approach considers web access sequences from the very beginning of web logs and therefore it is not suitable for mining data streams where the volume of data is huge and unbounded. At the same time, it cannot find the recent change of knowledge in data streams adaptively. The existing approach has many other limitations such as considering only forward references of web access sequences, suffers in the level‐wise candidate generation‐and‐test methodology, and needs several database scans, etc. In this paper, we propose a new approach for utility‐based web access sequence mining over data streams with a sliding window method. Our approach can not only handle large‐scale data but also efficiently discover the recently generated information from data streams. Moreover, it can solve the other limitations of the existing algorithms over data streams. Extensive performance analysis shows that our approach is very efficient and outperforms the existing algorithms.

국문 초록 (Abstract)

유틸리티 기반 웹 방문 패턴의 마이닝은 웹 페이지의 중요도, 사용자가 웹 페이지에 머문 시간을 유틸리티로정의하여 유티릴티 값이 큰 방문 패턴을 탐색하는 마이닝 기법을 말한다. 기존의 유틸리티 기반 웹 방문 패턴의 마이닝 기법들은 여러 번의 데이터베이스 스캔을 필요로 하는 것으로 데이터 스트림과 같은 웹 로그 데이터를 처리하기에는 적절하지 못하였다. 본 논문에서는 슬라이딩 윈도우 모델을 기반으로 한 번의 데이터베이스스캔을 통하여 유틸리티 기반 웹 방문 패턴을 탐색하는 기법을 제안한다. 제안하는 기법은 대용량의 데이터에대하서도 좋은 확장성을 보이고 슬라이딩 윈도우 모델을 통하여 최근의 정보를 탐색할 수 있는 기법임을 여러실험을 통하여 중명한다.

참고문헌 (Reference)

1 J. H. Chang, "estWin: Online data stream mining of recent frequent itemsets by sliding window method" 31 (31): 76-90, 2005

2 R. Agrawal, "ast algorithms for mining association rules" 487-499, 1994

3 B. Mobasher, "Web mining: Pattern discovery from World Wide Web transactions"

4 L. Zhou, "Utilitybased web path traversal pattern mining" 373-380, 2007

5 C. I. Ezeife, "Mining web log sequential patterns with position coded preorder linked wap‐tree" 10 (10): 5-38, 2005

6 R. Agrawal, "Mining sequential patterns" 3-14, 1995

7 H. Yao, "Mining itemset utilities from transaction databases" 59 (59): 603-626, 2006

8 J. Pei, "Mining access patterns efficiently from web logs" 396-407, 2000

9 Y.‐C. Li, "Isolated items discarding strategy for discovering high utility itemsets" 64 (64): 198-217, 2008

10 "IBM quest synthetic data generation code"

11 C. F. Ahmed, "Efficient tree structures for high utility pattern mining in incremental databases" 21 (21): 1708-1721, 2009

12 M.‐S. Chen, "Efficient data mining for path traversal patterns" 10 (10): 209-221, 1998

13 C. K.‐S. Leung, "DSTree: A tree structure for the mining of frequent sets from data streams" 928-932, 2006

14 H.‐F. Li, "DSM‐PLW: Single‐pass mining of path traversal patterns over streaming Web click‐sequences" 50 : 1474-1487, 2006

15 C. F. Ahmed, "An efficient candidate pruning technique for high utility pattern mining" 749-756, 2009

16 Y. Liu, "A two‐phase algorithm for fast discovery of high utility itemsets" 689-695, 2005

17 H. Yao, "A foundational approach to mining itemset utilities from databases" 482-486, 2004