MS-UCF: A reliable recommendation method based on mood-sensitivity identification and user credit

In recent years, due to an increasing overload of information on the Internet, there are many scenarios where recommendation systems are applied to provide suggestions to users. While prior works have made progress on addressing this challenge, an important limitation exists: they did not explore the mood-sensitivity aspect of the problem. The user's emotions affect the objectivity of the scoring; therefore the ratings are not the true reflection of user's preferences. Accordingly, recommendation systems make inaccurate recommendations for users. In this paper, MS-UCF based on the combination of mood-sensitivity (MS) and improved user-based collaborative filtering (UCF) is presented to improve the reliability of the recommendations. MS identifies the mood sensitivity of users by text analysis of their comments. This method is then combined with UCF to predict the preference of target users. User credit is also considered to resist fake users and make sure the recommendation reliable. Experiments are conducted to evaluate this approach. Compare it with other state-of-the-art approaches using the real-world movie dataset, the results show that with proposed approach, the recommendations are much reliable over comparative ones.