数据挖掘技术决策树分类算法分析、比较与实验  

Analysis,Comparison and Experiment of Classification Algorithm of Decision Tree in Data Mining Technology

作　　者：马俊宏[1] Ma Junhong(Jinzhong University, Jinzhong 030600, China)

机构地区：[1]晋中学院,晋中030600

出　　处：《北京印刷学院学报》2017年第7期159-160,186,共3页Journal of Beijing Institute of Graphic Communication

基　　金：<大数据工作室>;晋中学院"1331工程"重点创新团队建设计划资助科研课题

摘　　要：近些年来,互联网迅速发展,数据量每年都以惊人的幅度提升,人们的生活、政府的管理都和电子信息设备息息相关,特别是电子商务和科学实验数据库的迅速壮大,为我们带来了海量的数据。这些海量的数据中,往往蕴藏非常多有价值的记录和信息,等待着人们去挖掘,人们希望将这些信息分离提取出来进行更高程度的分析和统计,以便为我们所取用。而目前大部分数据库系统仅仅可以实现数据的增、删、改、查,很难找到大数据之间所蕴含的规则和关系,比较缺乏挖掘数据内部价值的有效方法,较难通过数据的维度去探索和发现、预测未来的趋势。本文通过对数据挖掘技术中决策树的分类算法做出实验分析,进行比较,给出合理的分析建议。In recent years,Internet has developed rapidly.The amount of data increases at an alarming range every year.People’s life and management of government are closely related to electronic information equipment.In particular,the rapid growth of e-commerce and scientific experiment database has brought us huge amounts of data.A lot of valuable records and information are stored in vast amounts of data,waiting for mining.People want to separate and extract these information for a higher level of analysis and statistics so that we can use them.However,at present,most database systems can only achieve data increasing,deleting,changing and checking,in which it is hard to find the rules and relationships between big data without effective ways to tap the internal value of data and it is difficult to explore,discover and predict future trend through the dimension of data.This paper makes an experimental analysis of the classification algorithm of decision tree in data mining technology,compares and gives reasonable analysis suggestions.

关 键 词：数据挖掘 决策树 ID3算法 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]