| 摘要: |
| 介绍了一种基于网络结构推荐的改进算法.在标准物质扩散算法的基础上，考虑到用户的评分对推荐商品的影响，对推荐算法中初始资源分配矢量和资源转移矩阵进行了改进，增加了调节因子.使用来源于GroupLens网站上的训练集数来评价这个推荐算法的性能，从而进行了一系列的实验.实验结果表明，该算法比传统的协同过滤系统、基于网络结构的推荐系统和带有权重的基于网络结构的推荐系统具有更好的推荐精度和更高的命中率，解决了标准物质扩散算法当中的冷启动问题和可扩展性问题，使得推荐结果具有多样性. |
| 关键词: 推荐算法 物质扩散算法 冷启动问题 可扩展性问题 推荐多样性 |
| DOI：10.3969/J.ISSN.1000-5137.2017.04.012 |
| 分类号:TP301.6 |
| 基金项目: |
|
| An improved recommended algorithm for network structure based on two partial graphs |
|
Deng Song, Qiu Jing, Chen Junhua
|
|
The College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai 200234, China
|
| Abstract: |
| In this thesis,we introduce an improved algorithm based on network structure.Based on the standard material diffusion algorithm,considering the influence of the user's score on the recommendation,the adjustment factor of the initial resource allocation vector and the resource transfer matrix in the recommendation algorithm is improved.Using the practical data set from GroupLens webite to evaluate the performance of the proposed algorithm,we performed a series of experiments.The experimental results reveal that it can yield better recommendation accuracy and has higher hitting rate than collaborative filtering,network-based inference.It can solve the problem of cold start and scalability in the standard material diffusion algorithm.And it also can make the recommendation results diversified. |
| Key words: recommendation algorithm standard material diffusion algorithm the problem of cold start the problem of capability recommended diversity |
