摘要
通过对决策树、k-Nearest Neighbor、贝叶斯三种不同数据挖掘算法的比较研究,基于可移动端数据的特点,建立了可移动端数据安全检测的模型框架,并通过实验对其加以验证。结果表明,决策树算法的检测分类结果最好,其查准率和查全率结果都很高;贝叶斯算法的检测分类结果性能稳定,但准确性不高,分类精度不理想,这是由该算法本身固有的特点决定的;k-Nearest Neighbor算法在开始时受到样本向量多少的影响,检测分类的效果不太稳定,分类效果在样本向量较少的情况下较差。通过对数据挖掘的可移动终端数据安全检测技术的研究,为今后数据安全检测技术的应用提供了一定的指导价值。
By comparatively studying on the data mining algorithms of decision tree, k-Nearest Neighbor and Bayesian, a model framework of the mobile terminal data security detection was established according to the characteristics of the mobile ter- minal data, and verified with the experiment. The results show that the decision tree algorithm has the best detection and classifi- cation result, and its precision ratio and recall ratio are both high; the Bayesian algorithm has the stable performance of the de- tection and classification result, but its accuracy is low and classification precision is unsatisfied because of the inherent charac- teristics of the algorithm itself; the k-Nearest Neighbor algorithm reflected by the quantity of the sample vectors has unstable de- tection and classification result, and the classification result is poor when the algorithm has less sample vectors. The mobile ter- minal data security detection technology of the data mining is studied, which provides a certain guidance value for the applica- tion of the data security detection technology.
出处
《现代电子技术》
北大核心
2017年第5期82-84,88,共4页
Modern Electronics Technique
基金
重庆市教育委员会"示范高职应用技术推广中心建设与社会服务能力创新研究(KJ122201)"的研究成果
关键词
数据挖掘
移动终端
数据安全
检测技术
data mining
mobile terminal
data security
detection technology
