{"title":"A Recommendation System Framework to Generalize AutoRec and Neural Collaborative Filtering","authors":"Ramin Raziperchikolaei, Young-joo Chung","doi":"10.1109/ICDMW58026.2022.00151","DOIUrl":null,"url":null,"abstract":"AutoRec and neural collaborative filtering (NCF) are two widely used neural network-based frameworks in the recommendation system literature. In this paper, we show that these two apparently very different frameworks have a lot in common. We propose a general neural network-based frame-work, which gives us flexibility in choosing elements in the input sources, prediction functions, etc. Then, we show that AutoRec and NCF are special forms of our generalized framework. In our experimental results, first, we compare different variants of NCF and Autorec. Then, we indicate that it is necessary to use our general framework since there is no specific structure that performs well in all datasets. Finally, we show that by choosing the right elements, our framework outperforms the state-of-the-art methods with complicated structures.","PeriodicalId":146687,"journal":{"name":"2022 IEEE International Conference on Data Mining Workshops (ICDMW)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2022-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 IEEE International Conference on Data Mining Workshops (ICDMW)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICDMW58026.2022.00151","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
AutoRec and neural collaborative filtering (NCF) are two widely used neural network-based frameworks in the recommendation system literature. In this paper, we show that these two apparently very different frameworks have a lot in common. We propose a general neural network-based frame-work, which gives us flexibility in choosing elements in the input sources, prediction functions, etc. Then, we show that AutoRec and NCF are special forms of our generalized framework. In our experimental results, first, we compare different variants of NCF and Autorec. Then, we indicate that it is necessary to use our general framework since there is no specific structure that performs well in all datasets. Finally, we show that by choosing the right elements, our framework outperforms the state-of-the-art methods with complicated structures.