摘要
论文主要研究了数据挖掘中基于信息熵的决策树的思想和算法,并针对"迷失无忧"这个基于移动互联网的商业运营平台遇到的UE问题,结合决策树算法的特点,提出了一种基于信息熵的决策树的预测模型。该模型利用训练数据集中的信息熵来形成分类,通过对模型的后台业务逻辑的应用来预测对未知状态下用户的使用习惯。该研究为互联网领域的大量类似问题提供了解决思路。
In this paper,the idea and algorithm of decision tree based on information entropy in data mining are studied.For the problem of UE based on mobile Internet and the characteristics of decision tree algorithm,a new model of decision tree based on information entropy is proposed.The model uses the information entropy of the training data set to form the classification.Through the application of the background business logic of the model,it is used to predict the user's usage habits.This study provides a solution to a large number of similar problems in the field of the Internet.
出处
《计算机与数字工程》
2016年第5期878-883,共6页
Computer & Digital Engineering
基金
2015年度广东大学生科技创新培育专项资金项目(编号:pdjh2015b0731)资助
关键词
决策树
预测
熵
用户体验
decision tree
prediction
entropy
UE