Analysis of Mood Tags for Multimedia Content Recommendation in Social Networks

The propensity of Web information purchasers is changing from cost effectiveness that emphasizes price with performance to cost satisfaction that emphasizes purchaser's psychological satisfaction. One of the methods to improve user's cost satisfaction in recommending multimedia contents is to use mood inherent in multimedia contents. SNS services based on mood folksonomy are applications using such a method. However, in the applications based on folksonomy, the problem caused by synonyms exists. This paper suggests a cost-satisfactive multimedia contents recommendation method to solve the problem on synonym. To this end, we utilizes the value of Arousal & Valence which express mood of multimedia content as its internal tag. A method that defines the relation between AV values of multimedia contents and AV values of mood tags is suggested by considering synonym and the correlation between them is analyzed. From the analysis, it is shown that the AV values of multimedia contents with a mood tag and its synonyms reside in the area of the mood tag in the Thayor model.