An item recommendation model with content semantic. (21st October 2019)

• Record Type: Journal Article
• Authors: Jiang, Yunpeng; Wang, Liejun; Qin, Jiwei
• Is Part Of: International journal of information and communication technology. Volume 15:Number 4(2019)
• Page Start: 370
• Page End: 390
• Publication Date: 2019-10-21
• Subjects: recommender model -- semantic feature -- similarities -- Word2vec -- data sparsity; Information technology -- Periodicals; Computer science -- Periodicals; Telecommunication -- Periodicals; 004.05
• Languages: English
• ISSNs: 1466-6642

Abstract

Current recommender service providers are offering interesting items for user-based user behaviour and ignoring the content semantic of items. The item semantic is should be taken into account as an accurate reflection of items. We present a recommender model that leverages content semantic and user rating. In this model, the item similarity is firstly calculated with content semantic by best Word2vec method, an item recommendation list is built by the similarities. Next, the user rating is used to model the user preference and build the other item list recommended by traditional recommendation method. Then, the two item lists is mixed together as final list for user. Comparing the above algorithm to traditional recommendation algorithms on MovieLens, FilmTrust and Online Retail datasets, we run experiments that show the presented algorithm has is greatly improved on accuracy and increase by an average of 25.32% to 31.41%, and present good scalability.