基于客户消费习惯的贝叶斯网络的分类算法  

Classification Algorithm of Bayesian Network Based on Customer Consumption Habits

作　　者：吕金锐[1] LV Jinrui(Taiyuan City Vocational College,Taiyuan Shanxi 030006)

机构地区：[1]太原城市职业技术学院,山西太原030006

出　　处：《软件》2024年第4期104-106,共3页Software

基　　金：山西省教育科学“十四五”规划课题“高职院校人工智能专业的建设”(GH-220096)。

摘　　要：面对大量用户的通信消费数据,如何对其进行挖掘从而获得有价值的信息,对客户进行分类,并制定不同的服务策略,是当前通信企业面临的一个普遍问题。本文介绍了几种常用的文本分类算法,通过分析用户的消费行为数据,选取了朴素贝叶斯分类预测算法、贝叶斯网络分类预测算法和决策树分类预测算法对用户进行分类,实验结果表明贝叶斯网络分类预测算法对于用户通信消费数据具有较好的分类效果。Faced with a large amount of communication consumption data from users,how to mine it to obtain valuable information,classify customers,and formulate different service strategies is a common problem faced by current communication enterprises.This paper introduces several commonly used text classification algorithms.By analyzing users'consumption behavior data,Naive Bayesian classification prediction algorithm,Bayesian Network classification prediction algorithm and Decision Tree classification prediction algorithm are selected to classify users.Experimental results show that Bayesian network classification prediction algorithm has good classification effect for users'communication consumption data.

关 键 词：文本分类 朴素贝叶斯 贝叶斯网络 决策树 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]