Mobile user exit feature selection method based on Fisher score and approximate Markov blanket

Abstract

The invention relates to a mobile user exit feature selection method based on an Fisher score and an approximate Markov blanket, and belongs to the field of data mining. Firstly, a Fisher criterion isutilized to reserve features with high classification capability, and irrelevant features and weak relevant features are eliminated; secondly, two measurement methods of a maximum information coefficient MIC and a symmetric uncertainty SU are fused, a correlation measurement standard MSCC is designed, and the MSCC standard is used for further removing irrelevant features; and finally, in combination with an MSCC measurement standard, redundant features in the Fisher candidate feature set are eliminated by utilizing an approximate Markov-Blanket judgment condition, and finally, an optimal feature subset with a relatively small dimension scale is obtained. According to the method, the exit features of the mobile users can be effectively selected, and the classification accuracy of the modelis improved.

Description

technical field [0001] The invention belongs to the technical field of data mining, and relates to a method for selecting a mobile user's outbound feature based on a Fisher score and an approximate Markov blanket. Background technique [0002] With the advent of the mobile Internet era, the number of mobile Internet users continues to increase, and people's life and work styles are quite different from before. The high penetration rate of mobile devices has brought about explosive growth in mobile data. Mobile data has the advantages of comprehensive data sampling and good real-time performance. It is quite authoritative in the field of trend analysis and potential user mining, and provides good convenient conditions for industry user mining. [0003] Feature selection is a key data preprocessing step in machine learning and data mining. It is the process of screening the most effective features from the original features to reduce the feature dimension of the data set, and ...

AI Technical Summary

Problems solved by technology

[0004] Mobile data generally contains high-dimensional features and is non-linear data. When the number of samples is limited, if a large number of features are used to design a classifier, the computational overhead is high, the classification performance is poor, and overfitting is prone to occur.

However, the above methods have the disadvantages of being easy to fall into local optimum and easy to delete useful features.

Images