E-MsNFIS: Efficient Negative Frequent Itemsets Mining Based on Multiple Minimum Supports

Abstract:

Negative frequent itemsets (NFIS) like (a₁a₂¬a₃a₄) have played important roles in real applications because we can mine valued negative association rules from them. In one of our previous work, we proposed a method, named e-NFIS to mine NFIS from positive frequent itemsets (PFIS). However, e-NFIS only uses single minimum support, which implicitly assumes that all items in the database are of the same nature or of similar frequencies in the database. This is often not the case in real-life applications. So a lot of methods to mine frequent itemsets with multiple minimum supports have been proposed. These methods allow users to assign different minimum supports to different items. But these methods only mine PFIS, doesn’t consider negative ones. So in this paper, we propose a new method, named e-msNFIS, to mine NFIS from PFIS based on multiple minimum supports. E-msNFIS contains three steps: 1) using existing methods to mine PFIS with multiple minimum supports; 2) using the same method in e-NFIS to generate NCIS from PFIS got in step 1; 3) calculating the support of these NCIS only using the support of PFIS and then getting NFIS. Experimental results show that the e-msNFIS is efficient.

• Periodical: Applied Mechanics and Materials (Volumes 411-414)
• Pages: 386-389
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.411-414.386
• Online since: September 2013
• Authors: Tian Tian Xu, Xiang Jun Dong
• Keywords: Multiple Minimum Supports, Negative Association Rule, Negative Frequent Itemsets, Positive Frequent Itemsets

References

[1] X. Wu, C. Zhang and S. Zhang: Efficient Mining of both Positive and Negative Association Rules, ACM Transactions on Information Systems, Vol. 22(2004), pp.381-405.

DOI: 10.1145/1010614.1010616

[2] X. -J. Dong , Z. -D. Niu, X. -L. Shi, X. -D. Zhang and D. -H. Zhu: Mining Both Positive and Negative Association Rules from Frequent and Infrequent Itemsets, Lecture Notes in Computer Science, Vol. 4632(2007), pp.122-133.

DOI: 10.1007/978-3-540-73871-8_13

[3] M.L. Antonie, O. Zaiane: Mining Positive and Negative Association Rules: An Approach for Confined Rules, PAKDD (2004), pp.27-38.

[4] Y. Zhang, C. Wang and Z. Xiong: Improved algorithm of mining association rules with negative items, Computer Engineering and Applications, Vol. 44(2008), pp.169-171.

[5] Y. Zhang, Z. Xiong, C. Wang and C. Liu: Study on association rules with negative items based on bit string, Control and Decision, Vol. 25(2010). pp.37-42.

[6] B. -G. Yuan and L. C: A novel mining algorithm for negative association rules. GCIS 2009, Vol. 2(2009), pp.553-556.

[7] X. -J. Dong, L. Ma and X. -Q. Han: E-NFIS: Efficient negative frequent itemsets mining only based on positive ones. ICCSN (2011), pp.517-519.

DOI: 10.1109/iccsn.2011.6013958

[8] T. -T. Xu and X. -J. Dong: Mining Frequent Patterns with Multiple Minimum Supports using Basic Apriori, FSKD (2013), in press.

DOI: 10.1109/icnc.2013.6818114

[9] X. -J. Dong, S. -J. Song, T. Han and Y. -C. Lu: Study on Negative Association Rules, Transactions of Beijing Institute of Technology, Vol. 24(2004), pp.978-981.

[10] B. Liu, W. Hsu and Y. Ma: Mining association rules with multiple minimum supports, KDD-99(1999).