In order to further improve the video recommendation effect of short video platform,a short video recommendation sys-tem is constructed based on collaborative filtering algorithm.Among them,singular value decomposition(SVD)and penalty are intro-duced to optimize the traditional collaborative filtering algorithm,and the performance of the algorithm is improved to some extent.The experimental results show that,compared with the traditional collaborative filtering algorithm,the improved collaborative filtering algorithm proposed in this study has better performance,and the RMSE value of the algorithm result is only 0.948;Compared with the traditional recommendation model,the short video recommendation model proposed in this study based on the improved collaborative filtering algorithm can more accurately predict the user's video score,and then can more accurately recommend short videos that meet the user's preferences.The above conclusions show that using the recommendation model proposed in this study can achieve better recommendation effect,and can recommend videos more in line with users'preferences,which has certain practical application val-ue.
关键词
混合液分离/系统控制/数据采集/介质传感器
Key words
short video platform/recommended algorithm/collaborative filtering algorithm
维普
万方数据