Improving the effectiveness of experiential decisions by recommendation systems
作者：
Highlights：
• We proposed a movie recommendation system (RS) integrating Self-Organizing Map (SOM) with a Neural Network System (NNS).
• We used the MovieLens 100 k database from the “GroupLens Research Project” of The University of Minnesota to test the RS.
• The database contained over 100 thousand ratings of 943 users on 1,682 movies during October 1997 to April 1998.
• The SOM produced 4 clusters of audience with high similarity and these clusters became the four nodes of the hidden layer in NNS.
• The RS provided richer user experiences and improved the accuracy of predicting movie ratings and the speed of data transfer.
摘要
•We proposed a movie recommendation system (RS) integrating Self-Organizing Map (SOM) with a Neural Network System (NNS).•We used the MovieLens 100 k database from the “GroupLens Research Project” of The University of Minnesota to test the RS.•The database contained over 100 thousand ratings of 943 users on 1,682 movies during October 1997 to April 1998.•The SOM produced 4 clusters of audience with high similarity and these clusters became the four nodes of the hidden layer in NNS.•The RS provided richer user experiences and improved the accuracy of predicting movie ratings and the speed of data transfer.
论文关键词：Recommendation system,Experiential decision,Multilayer perception model,Neural network system,Collaborative filtering system
论文评审过程：Available online 20 February 2014.
论文官网地址：https://doi.org/10.1016/j.eswa.2014.01.035
