基于主成分分析的支持向量机对购房意愿的分类研究  

The Classification of Purchasing houses based on Principal Component Analysis and Support Vector Machine

作　　者：邹玉梅[1] 范敬雅 张鹏程[1] 

机构地区：[1]山东科技大学数学与系统科学学院,山东青岛266590

出　　处：《技术与创新管理》2016年第5期544-546,共3页Technology and Innovation Management

基　　金：国家统计局重点课题(2014LZ41)

摘　　要：居民的购房意愿在整个宏观层面上影响着整个社会结构的变迁和转型。文中基于500份居民购房意愿调查问卷,利用主成分分析法提取了主要特征,对主成分序列建立了支持向量机分类模型。五折交叉验证结果表明:分类效果良好,对政府和房地产开发商进行客户细分、制定营销策略有一定的借鉴意义。The intention of purchasing houses affects the social structure and the transformation of the whole society. 500 questionnaires are cited in this paper, and then the main features are extracted through principal component analysis, and the model of support vector machine classification for principal component sequence is built. The numerical experiments show that the classification is fine, which has some reference value in dealing with customer segmentation and marketing strategy for the government and the. real estate developers.

关 键 词：购房意愿 主成分分析法 核函数 五折交叉验证 支持向量机分类 

分 类 号：TP891[自动化与计算机技术—检测技术与自动化装置]