The Research Of Recommend Algorithm Based On Context-Aware

In recent years,with the continuous improvement of the awareness ability of mobile devices,the recommendation system can obtain users' preferences and rich contextual information from the context log of mobile.The context information with the recommend system can enhance the accuracy of recommendation,to help users find the outcomes that are fit with the needs,Based on the latest research work,this paper puts forward two kinds of context-aware recommendation algorithm.The main work and innovation of this paper is reflected in the following aspects:1.In this paper,a new contextual recommendation algorithm,which combines environment context and item feature contexts,is proposed.we can build three latent models that represent the latent preferences of users toward contexts,the latent preference of items toward contexts,the users' preferences of items,and identify the latent preferences of users toward context,as well as users toward new and selected items.Finally,using the potential relationship between users and environment context to weighted the recommended score,we adjust the recommended list of items and recommend them to users.2.In the recommendation system,a critical challenge is that individual environment context log may not contain sufficient item access records for mining his/her environment context preferences.We designs a contextual topic-based relevance recommendation algorithm(CTRR).The CTRR algorithm uses the CTRR_LDA model and a postfiltering strategy to recommend items to users in a specific environment context.CTRR LDA is an improved LDA model,which combines environment context and item feature contexts to calculate the probability of the item appeared.3.The proposed algorithm is valid though designing experiment and analyzing recommended results on LDOS-CoMoDa data.Experimental results show that the improved context-aware recommendation algorithm has better recommendation than traditional context-aware recommendation algorithm.So it,to some extent,promote the progress of theoretical research and applied research of collaborative filtering algorithm.the reliability of algorithm is validated in the context-aware recommendation scenario.