CAJ | 학술논문

针对公交调查数据背后信息挖掘的需求，论文首先介绍了公交调查数据挖掘的理论概念及任务方法。概述了公交调查数据挖掘的常规流程，并且在分析了几种公交调查数据挖掘模型建模方法的基础上，得出 k-means模型最适宜对公交调查数据进行聚类分析。最后以达州市公交调查数据为样本实例，采用年龄、职业、每周乘坐公交天数、每月公交花费、偏好付款方式等几种属性，借以 SPSS Clementine 为软件平台、以 k-means 为模型对各属性数据进行聚类分析，软件运行后得到相似度较大的几组类别，根据不同样本含量的几组聚类进行图表分析，分别得出优化公共交通服务的相应建议，达到最初挖掘公交调查数据背后信息的目的。
침대공교조사수거배후신식알굴적수구，논문수선개소료공교조사수거알굴적이론개념급임무방법。개술료공교조사수거알굴적상규류정，병차재분석료궤충공교조사수거알굴모형건모방법적기출상，득출 k-means모형최괄의대공교조사수거진행취류분석。최후이체주시공교조사수거위양본실례，채용년령、직업、매주승좌공교천수、매월공교화비、편호부관방식등궤충속성，차이 SPSS Clementine 위연건평태、이 k-means 위모형대각속성수거진행취류분석，연건운행후득도상사도교대적궤조유별，근거불동양본함량적궤조취류진행도표분석，분별득출우화공공교통복무적상응건의，체도최초알굴공교조사수거배후신식적목적。
According to the need of mining information behind bus survey data,this paper mainly discusses the data mining technology in the application of bus survey data. First the paper introduces the theoretical concepts and tasks of data mining methods,summarizes the conventional process of data min-ing,and then on the bases of analysis the modeling methods of several data mining models,we find k-means is the most suitable model for bus survey data clustering analysis;paper also uses DaZhou bus sur-vey data for instances,with age,job,days per week,cost per month,type of payment,and uses SPSS Cle-mentine as a platform and k-means as a model for data clustering analysis,obtains larger similarity of sev-eral groups of categories,and carries on the chart analysis,get corresponding suggestions of optimizing bus service,achieve the objective of the initial information hidden in transit survey data.