Data Mining for Uncertain Data Based on Difference Degree of Concept Lattice
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Along with the rapid development of the database technology, as well as the widespread application of thedatabase management systems are more and more large. Now the data mining technology has already beenapplied in scientific research, financial investment, market marketing, insurance and medical health and so on,and obtains widespread application. We discuss data mining technology and analyze the questions of it. Therefore, the research in a new data mining method has important significance. Some literatures did notconsider the differences between attributes, leading to redundancy when constructing concept lattices. Thepaper proposes a new method of uncertain data mining based on the concept lattice of connotation differencedegree (c_diff). The method defines the two rules. The construction of a concept lattice can be accelerated byexcluding attributes with poor discriminative power from the process. There is also a new technique ofcalculating c_diff, which does not scan the full database on each layer, therefore reducing the number ofdatabase scans. The experimental outcomes present that the proposed method can save considerable time andimprove the accuracy of the data mining compared with U-Apriori algorithm.
Abstract
Along with the rapid development of the database technology, as well as the widespread application of thedatabase management systems are more and more large. Now the data mining technology has already beenapplied in scientific research, financial investment, market marketing, insurance and medical health and so on,and obtains widespread application. We discuss data mining technology and analyze the questions of it. Therefore, the research in a new data mining method has important significance. Some literatures did notconsider the differences between attributes, leading to redundancy when constructing concept lattices. Thepaper proposes a new method of uncertain data mining based on the concept lattice of connotation differencedegree (c_diff). The method defines the two rules. The construction of a concept lattice can be accelerated byexcluding attributes with poor discriminative power from the process. There is also a new technique ofcalculating c_diff, which does not scan the full database on each layer, therefore reducing the number ofdatabase scans. The experimental outcomes present that the proposed method can save considerable time andimprove the accuracy of the data mining compared with U-Apriori algorithm.
참고문헌 (18)
- M. Gao, "Data provenance management and similarity query over uncertain data", M. S., pp.2011
- S. Rao, "Implementing improved algorithm over APRIORI data mining association rule algorithm", International Journal of Computer Science and Technology(IJCST), 3(1), pp.489~493 (2012)
- J. Han, "Data Mining: Concepts and Techniques" (2001), Morgan Kaufmann
- H. Zhao, "Overview of association rules Apriori mining algorithm", Journal of Sichuan University of Science and Technology (Natural Science Edition), 24(1), pp.66~70 (2011)[DOI]
- L. Chi, "A two-class classification mining algorithm based on the FP-growth algorithm and its application", M. S., pp.2012
- Z. Liu, "Fast algorithm of frequent itemset mining based on matrix from uncertain data", Journal of Nanjing University of Science and Technology(Natural Science Edition), 39(4), pp.420~425 (2015)[DOI]
- J. Wang, "Survey on algorithm of mining frequent itemsets from uncertain data", Computer Engineering and Applications, 47(20), pp.121~125 (2011)[DOI]
- C. Chen, "Mining frequent items in uncertain dataset using compressed UF-tree", Application Research of Computers, 31(3), pp.716~719 (2014)[DOI]
- Z. Li, "An improved data mining algorithm based on concept lattice", Journal of Chongqing Normal University(Natural Science Edition), 30(2), pp.92~95 (2013)[DOI]
- J. C. W. Lin, "High average-utility sequential pattern mining based on uncertain databases", Knowledge and Information Systems, 62, pp.1199~1228 (2020)[DOI]
- Y. Baek, "Uncertainty-based pattern mining for maximizing profit of manufacturing plants with list structure", IEEE Transactions on Industrial Electronics, 67(11), pp.9914~9926 (2020)[DOI]
- M. M. Rahman, "Mining weighted frequent sequences in uncertain databases", Information Sciences, 479, pp.76~100 (2019)[DOI]
- G. Lee, "A new efficient approach for mining uncertain frequent patterns using minimum data structure without false positives", Future Generation Computer Systems, 68, pp.89~110 (2017)[DOI]
- G. Lee, "Analysis of tree-based uncertain frequent pattern mining techniques without pattern losses", The Journal of Supercomputing, 72, pp.4296~4318 (2016)[DOI]
- X. Zhao, "On relationship between three-way concept lattices", Information Sciences, 538, pp.396~414 (2020)[DOI]
- G. Liao, "Frequent pattern mining of uncertain data streams based on probability decay window model", Journal of Computer Research and Development, 49(5), pp.1105~1115 (2012)
- S. Wang, "Frequent item query algorithm based on uncertain data", Journal of Northeastern University(Natural Science Edition), 32(3), pp.344~347 (2011)[DOI]
- L. Zhang, "Theory and algorithms of attribute decrement for concept lattice", Journal of Computer Research and Development, 50(2), pp.248~259 (2013)