Title :
The research on optimization method for dynamic reduct
Author :
Wang Jia-yang ; Xue Shuang-ying ; Chen Si-ma
Author_Institution :
Coll. of Inf. Sci. & Eng., Central South Univ., Changsha, China
Abstract :
The paper analyses the deficiencies of traditional generalized dynamic reduct algorithm and proposes a fast algorithm based on the validity of reduct. The algorithm only needs to calculate part of the subtable to get the generalized dynamic reduct. It decreases the number of subtables that need to calculate the reduct by using the stability threshold, and filters the reduct of subtables that are not to be the generalized dynamic reduct by utilizing the optimal stability coefficient. The result shows that the fast method is more effective than the traditional algorithm.
Keywords :
data mining; optimisation; rough set theory; generalized dynamic reduct algorithm; knowledge discovery; optimal stability coefficient utilization; optimization method; reduct validity; rough set theory; stability threshold; Algorithm design and analysis; Educational institutions; Heuristic algorithms; NP-hard problem; Probability; Stability analysis; capacity of the family F; dynamic reduct; knowledge discovery; rough set; stability coefficient; stability threshold;
Conference_Titel :
Fuzzy Systems and Knowledge Discovery (FSKD), 2012 9th International Conference on
Conference_Location :
Sichuan
Print_ISBN :
978-1-4673-0025-4
DOI :
10.1109/FSKD.2012.6233767