基于Hadoop的电子商务个性化推荐算法——以电影推荐为例

以电影为推荐对象，选择了基于内容的推荐算法和基于协同过滤的推荐算法相结合的混合推荐算法，并在Hadoop平台下进行MapReduce并行化；通过线性组合的方式将两种推荐算法计算得到的电影相关性系数进行组合，实现本文系统的计算电影相关性的混合推荐算法，得到最终的电影相关性矩阵，构建电影关系网；最后对本文的提出的新型混合推荐算法工作模式进行了测试分析，证明改进方案具有良好的推荐精度。

E-Commerce Personalized Recommendation Algorithm Design Based on Hadoop: Taking Move Recommendations as an Example

This paper implements recommendation algorithm based on Item collaborative filtering recommendation algorithm and content-based parallelization, and uses the hybrid recommendation algorithm based on these two complementary recommendation algorithm for project-related calculations, and accesses network of the projects' relationships, namely the correlation coefficient matrix of each project. Then we do the relevant test analysis to the improved algorithm.