机器学习算法在用户行为中的应用

Machine Learning Algorithms in the Application of user Behavior

机器学习是人工智能相关领域中与算法相关的一个子域,是解决人工智能问题的一个途径,它允许计算机不停的模拟人的思考方式进行学习,来发掘出隐藏在数据背后的模型,并能对不完全信息进行推理,来构造新事物。目前机器学习的应用主要集中在数据挖掘,计算机视觉,自然语言处理,模式识别,搜索引擎等。文中将机器学习中的算法决策树C4.5,随机森林,贝叶斯网络应用到电商用户行为数据的挖掘中,解决用户行为属性与用户收入水平的分类情况问题;通过三种算法对用户行为的研究,得出决策树C4.5算法在用户收入分类上要优于后两种。

Machine learning is in the related fields of artificial intelligence related to the algorithm a child domain,is a way to solve the problem of artificial intelligence,it allows computer simulation human way of thinking in learning,to dig the hidden data model,and reasoning with incomplete information,and to construct the new things.The mainly concentrated in machine learning application are the data mining,computer vision,natural language processing,pattern recognition,search engines,etc.This paper,will be use the machine learning algorithm of decision tree C4.5,random forests and bayesian networks in electricity user's behavior data mining,solving the user behavior properties and classification of user levels of income;The study of user behavior through three algorithms,it is concluded that the decision tree C4.5 algorithm on user revenue classification is better than two.