Research on Construction of Key Personnel Judgment Model through Data Label

Abstract

Key personnel are important business objects in the management and control of public security elements. Through big data and machine learning technology, it is of great significance to model the years of police's experience on research and judgment for the actual combat and the construction of Public Security Prevention and Control System. Aiming at the problem of complicated data type and low quantization degree of public security data, which makes it inconvenient to use machine learning for model analysis directly, a method of constructing public security personnel label system is proposed in this paper to transform complex business data into quantized label features, and on this basis, a key personnel judgement model is constructed by using FP-Growth and XGBoost algorithm. Through comparative analysis, the research results show that the judgment model proposed in this paper is better than the traditional business integral model and logistic regression algorithm in Accuracy, Precision and Recall, which can better serve the actual police combat.