CAJ | 학술논문

针对人群分类管理问题，提出一种基于主元分析方法的新的单变量贡献度方法。首先，研究各行为属性对主元的贡献率，确定人群关键行为属性。然后，在确定关键属性的基础上，利用最大似然估计方法对人群进行分类管理。应用人群分类管理方法到警方侦破案件中，可大大减小案件的搜索范围，达到提高警方办案效率的目的。通过实例验证了方法的有效性。
침대인군분류관리문제，제출일충기우주원분석방법적신적단변량공헌도방법。수선，연구각행위속성대주원적공헌솔，학정인군관건행위속성。연후，재학정관건속성적기출상，이용최대사연고계방법대인군진행분류관리。응용인군분류관리방법도경방정파안건중，가대대감소안건적수색범위，체도제고경방판안효솔적목적。통과실례험증료방법적유효성。
A novel classification of the crowd is investigated based on principal component analysis method in this paper. First,the contribution rates provided by each behavior attributes for principal compo-nent are studied to determine the key behavioral attributes. Then,based on the identification of key attrib-utes,the residents are divided into several groups by the maximum likelihood estimator method. The pres-ented classification method of the crowd not only narrows the search areas,but also improves the efficien-cy of the police investigations. Finally,a case is given to show the effectiveness of the proposed method.