基于用户和内容的混合模式推荐算法研究

Research on Mixed Mode Recommendation Algorithm Based on User and Content

推荐系统中的协同过滤算法和基于内容推荐算法都能够进行相关推荐,但是算法中存在的不足会导致推荐结果不准确。为提高推荐准确度,本文提出一种混合模式的推荐算法,建立用户的兴趣模型,对目标用户进行个性化的物品推荐。最后利用Movies Lens数据集进行训练并评估基于用户和基于内容的混合模式推荐算法的准确度。

Both the collaborative filtering algorithm and the content-based recommendation algorithm in the recommendation system can make relevant recommendations,but the deficiencies in the algorithms can lead to inaccurate recommendation results.In order to improve the recommendation accuracy,this paper proposes a mixed-mode recommendation algorithm,establishes a user's interest model,and recommends personalized items to target users.Finally,the Movies Lens dataset is used to train and evaluate the accuracy of the user-based and contentbased hybrid mode recommendation algorithm.