| 一种双向聚类协同过滤推荐算法研究 |
| |
| 引用本文： | 范奥哲,何利力.一种双向聚类协同过滤推荐算法研究[J].教育技术导刊,2020,19(5):78-82. |
| |
| 作者姓名： | 范奥哲 何利力 |
| |
| 作者单位： | 浙江理工大学 信息学院，浙江 杭州 310018 |
| |
| 基金项目： | 浙江省科技厅（重大）项目（2015C03001） |
| |
| 摘 要： | 针对传统协同过滤推荐算法在大数据环境下存在数据稀疏性及计算复杂性等问题，提出一种双向聚类协同过滤推荐算法。该算法首先从用户维度和项目维度两个方向分别进行属性聚类，然后在目标用户和目标项目所在类簇中分别使用改进后的相似度计算方法进行协同过滤推荐，最后通过平衡因子综合预测评分并形成最终推荐列表。在 MovieLens 公开数据集上进行实验，结果表明，该算法（DCF）相比传统协同过滤推荐算法（TCF）、基于用户聚类的协同过滤推荐算法（UCF）以及基于项目聚类的协同过滤推荐算法（ICF），在平均绝对误差上分别降低了 16%、8.1%、7.5%，有效提高了推荐精度。
|
| 关 键 词： | 协同过滤推荐算法 数据稀疏性 聚类 推荐系统 |
| 收稿时间： | 2019-07-08 |
A Collaborative Filtering Recommendation Algorithm for Bidirectional Clustering |
| |
| FAN Ao-zhe,HE Li-li.A Collaborative Filtering Recommendation Algorithm for Bidirectional Clustering[J].Introduction of Educational Technology,2020,19(5):78-82. |
| |
| Authors: | FAN Ao-zhe HE Li-li |
| |
| Institution: | School of Information Science and Technology，Zhejiang Sci-Tech University，Hangzhou 310018，China |
| |
| Abstract: | In order to solve the problems of traditional collaborative filtering recommendation algorithm in data sparsity and computational complexity in big data environment，a collaborative filtering recommendation algorithm based on bidirectional clustering is proposed. The algorithm firstly performs attribute clustering from the user dimension and the project dimension，and then uses the im?
proved similarity calculation method to perform collaborative filtering recommendation in the cluster of the target user and the target project respectively. Finally，the comprehensive prediction by balance factor is used and the final list of recommendations is formed. Experiments on the MovieLens public dataset show that the proposed algorithm（DCF）is more efficient than the traditional collaborative filtering recommendation algorithm（TCF），user clustering based collaborative filtering recommendation algorithm（UCF），and project clustering based collaborative filtering recommendation algorithm（ICF）. The proposed algorithm can reduce the average absolute error by 16%，8.1% and 7.5%，respectively，and the algorithm improved the accuracy of the recommendation. |
| |
| Keywords: | collaborative filtering recommendation algorithm data sparsity clustering |
|
| 点击此处可从《教育技术导刊》浏览原始摘要信息 |
| 点击此处可从《教育技术导刊》下载免费的PDF全文 |
