Research On Recommendation Method Based On Context-Aware And Multi-Context Fusion

Over time,with the rapid development of big data,how to solve the related "information overload" problem has become increasingly apparent.Accurate and rapid access to Internet information has become a hot research direction for scholars.The emergence of search engines can filter a large amount of information based on user preferences and provide them with useful elements.The proposed system can filter a large amount of information based on user preferences and provide them with useful elements.The formation of a recommendation system effectively alleviates the problem of massive data filtering,filtering unimportant information based on users’ interest preferences,and recommending items of interest.It is considered a powerful tool that can help businesses promote sales and also help users make decisions.Aiming at the problem of neglecting the influence of user emotion on user behavior in traditional recommendation systems and not fully considering user item preference in recommendation systems based on diffusion principle,based on the emotional analysis of user comments and the comprehensive score of user items,this paper proposes a weighted hybrid recommendation method combining material diffusion and heat conduction with user emotion.Based on the analysis of user comments,the user emotion is modeled.Integrate user emotion tags and build a weighted three-part graph model recommendation method.This method,on the one hand,makes full use of the principle of material diffusion to improve the accuracy of recommendation,on the other hand,uses the heat conduction theory to improve the diversity of recommendation objects,and balances the accuracy and diversity through resource integration,and finally realizes effective personalized recommendation of emotional perception.On two real data sets,it is verified that the recommendation performance of this model is superior to other similar recommendation models.Compared with the traditional recommender systems,context-aware recommender systems are more in line with actual application contexts.However,the existing researches are mostly focused on single context-aware recommendation,such as time-aware recommendation or location-aware recommendation,and lack of in-depth research on multi-context-aware recommendation.Therefore,we proposed a recommendation method of high-order tensor factorization based on multi-context-aware.Firstly,on the basis of analyzing the influence of context on users’ interest preferences,the sensitivity of users to multiple contexts was detected by using statistical methods.For context-sensitive users,four-dimensional tensors and feature matrices used to solve data sparsity were constructed based on rating matrix and situational information.And then the stochastic gradient descent algorithm was used for iterative calculation to fill in missing data values and carry out parameter optimization.For context-insensitive users,we used matrix factorization to predict users’ interest preferences.Finally,we tested and validated our method on a multi-context-aware movie dataset,and the experimental results show that the proposed method could effectively reduce the prediction error and improve the recommendation quality.