基于用户行为的网站推荐系统模型  

Recommended System Model Based on User Behavior

作　　者：吴俊杰[1] 刘耀军[1] 赵月爱[1] 

机构地区：[1]太原师范学院计算机系,山西太原030012

出　　处：《山西电子技术》2011年第6期66-67,共2页Shanxi Electronic Technology

基　　金：山西省软科学研究基金资助项目(2010041058-04);山西省科学技术发展计划基金资助项目(20110321024-02)

摘　　要：网上营销的难点是如何寻找潜在客户。识别用户行为是一个复杂的过程,建议采用二维模型来寻找相关的项目,然后推荐给用户。但太多不相干的建议转化为用户信息过载问题。本研究的目的是在每个维度中找出已被用户搜索的项目/对象,最受好评的每个维度中的项目可给予建议。要做到这一点,我们采用张量,它可以有效地模拟多维Web日志中的用户搜索数据,然后确定在每一个搜索维度上的用户模型。所有的评估均基于真实数据集,这清晰地显示出比其他广泛使用的向量和矩阵方法的有效性。How to find potential customers is the difficulty of online marketing. The paper recommends the using of two-dimensional model to find related items, and then recommend to the user. But too many irrelevant suggestions are translated into the overload problem of user information. The purpose of this study is to identify in each dimension the project / object that searched by users, and the most award-winning project in each dimension can give advice to user. The paper uses tensor to do this work, which can effectively simulate the multi-dimensional Web logs in user＇ s search data, then determine a search of user model in each dimension. All assessments are based on real data sets, which dearly show the effect than the widely used vector and matrix method.

关 键 词：用户行为建模 张量 矩阵分解 推荐 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]