基于ART算法的电子商务个性化聚类模型的设计与实现

E-COMMERCE CHARACTERISTIC-CLUSTERING MODEL BASED ON ART ALGORITHM: DESIGN AND ITS IMPLEMENTATION

本文针对个性化电子商务网站建设中出现的难于有效发现用户行为特征问题 ,提出一种基于 ART神经网络自适应谐振算法的个性矢量聚类模型 .该聚类模型由两个智能子系统和三个逻辑控制单元组成 ,采用二值输入模式 ,具备很强的自适应性 .模型实现是在 C+ +语言平台上进行的 .在模型程序设计中 ,采用衍生类方式构造子系统单元 ,通过控制对话、数据共享建立系统单元之间的联系 .该模型可以有效挖掘网络用户行为典型个性特征 ,用于指导电子商务网站资源的组织和再分配 .

This paper suggests a characteristic clustering model based on ART algorithm, which contraposes some difficulties of finding typical characteristic of users .The clustering model consists of two intelligent subsystems and three logical units, adopting two value inputs, having powerful capacity of self adapting. The models implementation employs C++ language platform .In the process of programming, the model introduces original class to build subsystem and establishes relations via controlling dialogue and sharing DATA. This model mines typical characteristic of actions of users of Internet, farther it can be used to direct organizing and re allotting resources of E commence sites.