摘要
全球每天产生大量的数据,如何快速处理大数据为人们所用是亟待解决的问题。随着大数据和数据挖掘技术的不断发展和成熟,个性化推荐在我们生活中发挥着越来越重要的作用。论文主要研究同一用户对相似电影以及相似用户对同一影片的评分预测,通过寻找最优参数的方法使预测的准确度提高,以便能够有效地向用户推荐其感兴趣的影片。论文利用MovieLens数据集,在Spark平台架构上训练得到最优ALS模型参数,并将模型预测出的评分与均值模型预测评分做对比。实验结果表明该文模型的预测准确度有了明显提高。
The world produces a lot of data every day,how to quickly deal with large data for people to use is an urgent problem to be solved. With the large data and data mining technology continues to develop and mature,personalized recommendation in our lives plays an increasingly important role. This paper focuses on the same user's prediction of the same film for similar films and similar users,and improves the accuracy of the forecast by finding the optimal parameters so that the videos of interest can be effectively recommended to the users. In this paper,the optimal ALS model parameters are trained on the Spark platform using the MovieLens data set,and the predicted scores of the model are compared with the mean model predictions. The experimental results show that the prediction accuracy of the model has been improved obviously.
作者
姜婷婷
杜振军
JIANG Tingting;DU Zhenjun(Dalian Maritime University School,Dalian 116026)
出处
《计算机与数字工程》
2018年第11期2310-2314,共5页
Computer & Digital Engineering