Data Mining Analysis for Improving Decision-Making in Computer Management Information Systems

Main Article Content

Xiaohong Dong
Bing Xiang


In this paper a decision tree-based data mining procedure for information systems is proposed to enhance the accuracy and efficiency of data mining. An enhanced C4.5 decision tree method based on cosine similarity is suggested to evaluate the information gain rate of characteristics and the information entropy of their values. When the information entropy variance among any two values for attributes is within the threshold range, the cosine similarity of the merging attribute values is determined, and the information gain rate of the attributes is recalculated. Large-scale data sets that conventional data processing methods are unable to handle successfully have given rise to the area of data mining. The prime objective is to look into how data mining technology is used in computer management information systems. The benefits of data mining technologies in computer management information systems are examined from a variety of angles in this study. In order to analyze and comprehend huge data sets and to derive knowledge that can be utilized to enhance the decision-making process in computer management information systems, the suggested solution makes use of a number of data mining techniques, including Clustering, Classification, and Association Rule Mining. The experimental analysis indicates that the time required by the proposed method to construct a decision tree is less than the time required by the GBDT, P-GBDT method and the C5.0 decision tree Hyperion image forest type fine classification method. The minimum time is not more than 15 seconds when compared with the minimum time saving of the other two methods. The time required by the C5.0 decision tree Hyperion image forest type fine classification method is always the greatest in comparison with the minimum time saving of the C5.0 decision tree. The classification accuracy of the proposed method for various datasets exceeds 95 percent, and the data mining efficacy is high. This method enhances the precision and efficacy of data mining in order to uncover valuable information concealed behind a large volume of data and maximize its value.

Article Details

Special Issue - Intelligent Cloud Technologies