基于自组织神经网络与主成分分析的多维度网络成瘾研究  

Internet addiction research based on the self-organizing neural network and multidimensional principal component analysis

作　　者：侯其锋[1] 孙兆恒[2] 杨锆[3] 刘加海[3] 

机构地区：[1]西安电子科技大学数学与统计学院,陕西西安710071 [2]哈尔滨工程大学计算机科学与技术学院,黑龙江哈尔滨150001 [3]浙江大学城市学院,浙江杭州310015

出　　处：《浙江大学学报（理学版）》2015年第4期499-504,共6页Journal of Zhejiang University（Science Edition）

基　　金：陕西省社会科学基金资助项目(2014A05);中央高校基本业务费项目(7214517401);江苏省博士后科研资助计划首批资助项目(1401133C)

摘　　要：应用陈淑惠编制的中文网络成瘾量表对分别座落于我国东部和西部的2所高校的420名学生进行了问卷调查,采用自组织神经网络与主成分分析技术,对测试样本的网络成瘾总分进行聚类分析,并探求了网络成瘾的主要影响因素.结果表明:(1)上述方法能有效利用样本数据之间的相似性,发现其内在本质联系;(2)能够对测试样本进行更为有效的分类;(3)发现时间管理能力与强迫性上网行为等是造成网络成瘾的决定性因素.自组织神经网络对网络成瘾程度有更好的区分度,可以在宏观和微观2个层面对网络成瘾提出针对性的防范措施.The internet addiction data of 420 students from two universities in the east and west of China are obtained by the Chinese internet addiction scale（CIAS）compiled by CHEN.Based on the above data,self-organizing neural network and principal component analysis have been used respectively to make clustering analysis for the internet addiction disorder（IAD）samples and to find out the main impact factors of IAD.The results are as follows：（1）The above method can effectively utilize the similarity between the test samples and find their essential contacts.（2）The method can classify the test samples more effectively.（3）Time management ability and compulsive internet use play decisive roles in IAD.The conclusion is that the self-organizing neural network can distinguish the degree of internet addiction better.And,we can put forward two pointed preventive measures for IAD at both macroscopic and microcosmic levels.

关 键 词：网络成瘾 自组织神经网络 主成分分析 

分 类 号：B849[哲学宗教—应用心理学]