论文部分内容阅读
提出了一种新的信息系统属性约简算法。为此,首先建立了信息系统与关系矩阵之间的联系;其次,从关系矩阵的角度研究了合理刻画属性重要性的新指标;然后利用新指标作为启发式信息设计了一种新的属性约简算法。与现有算法相比,该算法具有较大的灵活性,它能从搜索空间中逐次删除不重要属性,避免对其重要性的重复计算。此外,对该算法的时间复杂度进行了详细的分析,并通过实例和实验验证它的可行性与有效性。
A new information system attribute reduction algorithm is proposed. To this end, we first establish the relationship between information system and relation matrix. Secondly, we study the new index which reasonably describes the importance of attributes from the perspective of relational matrix. Then, we design a new attribute about heuristic information Jane algorithm. Compared with the existing algorithms, this algorithm has more flexibility. It can delete the less important attributes from the search space one by one, and avoid double counting of its importance. In addition, the time complexity of this algorithm is analyzed in detail, and its feasibility and effectiveness are verified through examples and experiments.