【摘要】 在当今社会,手机犯罪越来越引起人们的重视,对研究人员来说需要马上研究相应的对策加以应对,智能手机的使用率越来越高也促使了手机取证技术的研究发展。同时采用Android系统的智能手机越来越多,针对Android系统手机取证的电子证据进行相应的数据分析,能更方便和直观的发现手机信息中的重点与需要关注的目标对象。在本文中应用了朴素贝叶斯分类算法对数据中各联系人进行分类,而朴素贝叶斯分类算法的条件独立性假设是非常苛刻的,很难在正常情况下满足,本文中提出了一种基于变异系数法的加权朴素贝叶斯分类模型,克服这个问题关键在于利用各项指标间所包含的信息的差异,通过计算得到指标的权重。有效地提高了朴素贝叶斯算法的分类性能,并且也继承了贝叶斯分类算法的简单性,本文首先对算法原理进行了分析与证明,然后描述了相应的算法,在最后给出了基于变异系数法的属性权值求解方法。更多还原

【Abstract】 In modern society,mobile phone crime phenomenon as a high-technology crime,need to study and the corresponding counter-measures to deal with,the popularity of intelligent mobile phone make the mobile phone on evidence research to a new height,wherein more and more intelligent mobile phone use Android system.this paper mainly introduces the Android system mobile phone forensics electronic sources of evidence and forensic analysis method,finally puts forward using Android system should solve the problem of mobile phone forensics. Naive Bayes is based on an assumption of conditional independence and the assumption can scarcely be satisfied. A weighted naive Bayes classification algorithm based on Coefficient of Variation is proposed. By computing Coefficient of Variation between condition attributes and decision attribute,different condition attributes are weighted differently. With a new method offered first to solve the weights of attributes on the basis of Coefficient of Variation discusses the operation principle of the algorithm,as well as its implementation.更多还原