LAUSR

With the rapid development of mobile Internet applications, it is more and more difficult for mobile users to choose the right commodity in the face of a large amount of commodity information. Recommendation system is one of the effective means to solve the problem of information overload. However, the low-hardware configuration and limited computing capacity are common problems in most mobile devices, which make mobile computing applications face more and more serious challenges with increasingly strong demand for computing resources. Cloud computing, as a model of providing resources on demand, can provide technical support for solving the above problems. Recommendation as a service (RaaS), which packaged the recommendation system as a service, has attracted extensive attention from the business community and academia. In this paper, the RaaS model based on emotion analysis and double labeling is proposed. In this model, we make full use of commodity description information and user comment text to explore users’ emotional preference on the commodity attributes that is less considered in the existing recommendation models or algorithms. First, we extract the keywords by analyzing the commodity description and user behaviors to establish commodity intrinsic attribute labels and user preference labels. Second, we mine the user’s comments through emotion analysis to establish the commodity feedback labels and user concern labels. Finally, the recommendation results for users are generated by calculating the similarity of above four types of labels. Experimental results show that the proposed model and its algorithms can improve the effect of recommendation.