Probabilistic Models for Local Patterns Analysis
2
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
Recently, many large organizations have multiple data sources (MDS’)distributed over different branches of an interstate company. Local patterns analysishas become an effective strategy for MDS mining in national and internationalorganizations. It consists of mining different datasets in order to obtain frequentpatterns, which are forwarded to a centralized place for global pattern analysis. Varioussynthesizing models [2,3,4,5,6,7,8,26] have been proposed to build global patternsfrom the forwarded patterns. It is desired that the synthesized rules from suchforwarded patterns must closely match with the mono-mining results (i.e., the resultsthat would be obtained if all of the databases are put together and mining has beendone). When the pattern is present in the site, but fails to satisfy the minimum supportthreshold value, it is not allowed to take part in the pattern synthesizing process. Therefore, this process can lose some interesting patterns, which can help the deciderto make the right decision. In such situations we propose the application of aprobabilistic model in the synthesizing process. An adequate choice for a probabilisticmodel can improve the quality of patterns that have been discovered. In this paper, weperform a comprehensive study on various probabilistic models that can be applied inthe synthesizing process and we choose and improve one of them that works toameliorate the synthesizing results. Finally, some experiments are presented in publicdatabase in order to improve the efficiency of our proposed synthesizing method.
Abstract
Recently, many large organizations have multiple data sources (MDS’)distributed over different branches of an interstate company. Local patterns analysishas become an effective strategy for MDS mining in national and internationalorganizations. It consists of mining different datasets in order to obtain frequentpatterns, which are forwarded to a centralized place for global pattern analysis. Varioussynthesizing models [2,3,4,5,6,7,8,26] have been proposed to build global patternsfrom the forwarded patterns. It is desired that the synthesized rules from suchforwarded patterns must closely match with the mono-mining results (i.e., the resultsthat would be obtained if all of the databases are put together and mining has beendone). When the pattern is present in the site, but fails to satisfy the minimum supportthreshold value, it is not allowed to take part in the pattern synthesizing process. Therefore, this process can lose some interesting patterns, which can help the deciderto make the right decision. In such situations we propose the application of aprobabilistic model in the synthesizing process. An adequate choice for a probabilisticmodel can improve the quality of patterns that have been discovered. In this paper, weperform a comprehensive study on various probabilistic models that can be applied inthe synthesizing process and we choose and improve one of them that works toameliorate the synthesizing results. Finally, some experiments are presented in publicdatabase in order to improve the efficiency of our proposed synthesizing method.
참고문헌 (27)
- Thirunavukkarasu Ramkumar, "The Effect of Correction Factor in Synthesizing Global Rules in a Multi-databases Mining Scenario", Journal of computer science, 6(3) (2009)
- Xindong Wu, "Synthesizing High-Frequency Rules from Different Data Sources", IEEE TransKnowledge Data Eng, 15(2), pp.353~367 (2003)
- Thirunavukkarasu Ramkumar, "Modified algorithms for synthesizing high-frequency rules from different data sources" (2008), Springer-Verlag
- Thirunavukkarasu Ramkumar, "Multi-Level Synthesis of Frequent Rules from Different Data Sources", ~ (2010)
- Shichao Zhang, "Mining globally interesting patterns from multiple databases using kernel estimation" (2009)
- Shichao Zhang, "Mining Multiple Data Sources: Local Pattern Analysis", Data Mining Knowledge Discovery, 12(2–3), pp.121~125 (2006)
- Animesh Adhikari, "Synthesizing heavy association rules from different real data sources", Elsevier Science Direct (2008)
- Animesh Adhikari, "Developing Multi-databases Mining Applications" (2010), Springer-Verlag
- Beyond independence: Probabilistic models for query approximation on binary transaction data, "D. Pavlov", Information and Computer Science, University of California, ~, pp.Technical Report UCI-ICS-TR-01-09 (2001)
- V. Poosala, "Selectivity estimation without the attribute value independence assumption", Proceedings of the 23rd International Conference on Very Large Data Bases (VLDB’97), pp.486~495 (1997), Morgan Kaufmann Publishers
- J. N. Darroch, "Generalized iterative scaling for log-linear models", Annals of Mathematical Statistics, 43, pp.1470~1480 (1972)
- Toon Calders, "Quick Inclusion-Exclusion" (2006), Springer-Verlag
- Unil Yun, "Efficient mining of weighted interesting patterns with a strong weight and", or support affinity, 2007, pp.177, 3477~3499
- S. Jaroszewiez, "An inclusion-exclusion result for Boolean polynomial and its applications in data mining", proceedings of the Discrete Mathematics in DM Workshop.SIAM DM Conference (2002)
- Szymon Jaroszewiez, "Support Approximation Using Bonferroni Inequalities", University of Massachusetts, ~
- A.L. Berger, "A maximum entropy approach to natural language processing", Computational Linguistics, 22(1), pp.39~72 (1996)
- J. Pearl, "Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference" (1988), Morgan Kaufmann Publishers Inc.
- F. Jelinek, "Statistical Methods for Speech Recognition" (1998), MTT Press
- Probabilistic models for query approximation with large Sparse Binary Data Set, "D. Pavlov", department of information and computer science, University of California, ~, pp.Technical Report N°:00-07 (2000)
- CHAIB Souleyman, "Fouille multi-bases de données : Analyse des motifs locaux", 석사, pp.2011
- H. Mannila, "Multiple uses of frequent sets and condensed representations", Proc. KDD Int. Conf. Knowledge Discovery in Databases (1996)
- H. Mannila, "Predictions with local patterns using cross-entropy", Proceedings of Fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (SIGKDD’99), pp.357~361 (1999), ACM Press
- C. K. Chow, "Approximating discrete probability distributions with dependence trees", IEEE transactions on Information Theory, IT-14(3), pp.462~467 (1968)
- Rakesh Agrawal, "Fast Algorithms for Mining Association Rules", VLDB’94, pp.487~499 (1994)
- Jiawi Han, "Mining Frequent Patterns without Candidate Generation:A Frequent-Pattern Tree Approach", Data Mining and Knowledge Discovery, 8, pp.53~87 (2000)
- Efficient mining of association rules using closed itemSet lattices, "N. PASQUIER", Information Systems, 25~46 (1999)
- Zhang S, "Multi-databases mining", IEEE Comput Intell Bull, 2, pp.5~13 (2003)