INFORMATION ON DOCTORAL THESIS
1. Full name: CU THU THUY 2. Sex: Female
3. Date of birth: 18/12/1972 4. Place of birth: Phu Tho
5. Admission decision number: 2377 Dated: 14/12/2006
6. Changes in academic process:
7. Official thesis title: Researching to mine the rare association rules and their application
8. Major: Information System 9. Code: 62 48 05 01
10. Supervisors:
Assoc. Prof. Dr. Do Van Thanh
Assoc. Prof. Dr. Ha Quang Thuy
11. Summary of the new contributions of the thesis:
About the theoretical research, the thesis has proposed to develop some kinds of sporadic rules and has developed algorithms to find rare itemsets for these rules.
On the transactional database, the thesis has been done on the approach to find out the closed rare itemsets for the rare rules instead of finding all of the rare itemsets like the previous researches. The thesis has been proposed three algorithms mining rare itemsets for three kinds of rare rules such as MCPSI algorithm mining closed perfectly sporadic itemsets with two thresholds, MCISI algorithm mining closed imperfectly sporadic itemsets with two thresholds and NC-CHARM algorithm mining the itemsets with negative items constraint.
For the problem of mining rare rules on quantitative database, the thesis has used the theory of fuzzy set to transfer the quantitative database into fuzzy database and to discover the rare rules relied on this fuzzy database. The thesis has proposed MFPSI algorithm to mine the fuzzy perfectly sporadic rules with two thresholds and MFISI to discover the fuzzy imperfectly sporadic rules with two thresholds.
12. Practical applicability, if any:
The thesis has proposed the combination between the mining negative association rules and smooth transition regression model on the construction of analytic and predicative model. The thesis has built a predicative model of securities index and a predicative model of Vietnam Consumption Price Index (CPI) based on this approach. Those models do not only help to analyze the economic relations, especially the cause and effect relation, but improve highly the quality of predication, accuracy of the predication also.
13. Further research directions, if any:
Our future research will be:
a) Finding the solutions to generate useful sporadic association rules from sporadic itemsets.
b) Applied approach parallel mining rare association rules.
c) Continue application of the mining association rules combined with other methods to analyze and forecast economic data.
14. Thesis-related publications:
[1]. Cu Thu Thuy, Do Van Thanh (2008), “A New Solution for Analyzing Vietnamese Stock Market”, Journal of Computer and Cybernetics, Vol 24 (2), tr. 107-118.
[2]. Cu Thu Thuy, Do Van Thanh (2009), “Mining Association Rules with Negative Item Constraints”, Journal of Computer and Cybernetics, Vol 25 (4), pp. 345-354.
[3]. Do Van Thanh, Pham Thi Thu Trang, Cu Thu Thuy (2009), “Building Price Forecasting Model by Combining the Smooth Transition Regression Model and Mining Association Rules”, In Proceedings of the second workshop in the framework of the Protocol Vietnam – Thailand, University of National Economy, pp 308-322.
[4]. Cu Thu Thuy, Do Van Thanh (2010), “Mining Perfectly Sporadic Rules with Two Thresholds”, In Proceedings of MASS2010, Wuhan, China.
[5]. Cu Thu Thuy, Do Van Thanh (2010), “Mining Imperfectly Sporadic Rules with Two Thresholds”, International Journal of Computer Theory and Engineering, Vol. 2 (5), pp. 1793-8201.
[6]. Cu Thu Thuy, Ha Quang Thuy (2010), “Mining Fuzzy Perfectly Sporadic Association Rules with Two Thresholds”, In Proceedings of the 13th National Conference on Information Technology and Communications, Hung Yen, pp 263-275.
[7]. Do Van Thanh, Cu Thu Thuy, Pham Thi Thu Trang (2010), “Building CPI Forecasting Model by Combining the Smooth Transition Regression Model and Mining Association Rules.”, Journal on Information Technologies and Communications, Vol E-1 (7), pp.16-27.
[8]. Cu Thu Thuy, Ha Quang Thuy (2011), “ Mining Fuzzy Imperfectly Sporadic Itemsets with Two Thresholds”, Journal of Computer and Cybernetics, Vol 27 (2), tr. 142-153.