Parallel algorithm of attributes reduction of decision table based on conditional entropy
一种基于条件熵的决策表属性约简并行算法*

This paper propose a parallel algorithm of attributes reduction of decision table based on conditional entropy. Through the calculation of the entropy, this algorithm which is on basis of classification is reducing the attributes while it is disassembling the decision table into sub-tables as symmetrical as possible correspondingly, so the algorithm is achieve a parallel calculation of attribute reduction. This paper also analyzes the time complexity of the algorithm and in the end the experiment shows that the new algorithm is more effective than traditional algorithm.