An improved heuristic attribute reduction algorithm based on the attribute frequency is presented. After analyzing many other attribute reduction algorithms, we utilize the discernibility matrix and the appeared attribute frequencies to determine each attribute?s significance, based on the principle of maximum attribute frequency,we achieved the reduction of the information system. An illustrative example demonstrate the algorithm?s effectiveness and validity.
Citation:
Haijun Wang, Shaoliang Wei, Yimin Chen, "An Improved Attribute Reduction Algorithm Based on Rough Set," snpd, vol. 3, pp.1007-1010, Eighth ACIS International Conference on Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing (SNPD 2007), 2007