Xgboost Algorithm Based Research and Modeling of Mate Selection Psychology of Highly Educated Female

Abstract

The mate selection tendency of highly educated women has both individuality and generality. In this paper, the mate selection tendency of highly educated women is constructed as a mathematical problem of binary classification prediction, and a scoring function is given to evaluate the prediction model based on certain assumptions. Based on the real data of an Internet dating platform in China, this paper extracts the basic attributes and socio-economic attributes of men to form an independent variable set, and proposes a prediction model of mate selection tendency of highly educated women based on Xgboost algorithm. The model achieves good prediction performance in both training data sets and test data sets. The results of the model in the test data set show that its normalized income is 82.9%, and the success rate of recommendation is about 72.7%, 2.23 times that of random recommendation, which can be applied to the spouse selection accurate recommendation function of the marriage platform.