UGC quality evaluation based on meta-learning and content feature analysis

With the fast development of Social Networking Services, there has been increasingly vast amount of information published by massive network users. Given this information explosion, how to analyze the quality of User Generated Contents (UGC) automatically becomes a challenging task for researchers. To solve the problem, we need to build an effective UGC quality evaluation system. In the light of our experience, we believe that the textual content of UGC is the key factor for its quality. Hence, we focus on textual content based quality evaluation and classification instead of using UGC publishing related data, such as times being commented and forwarded in this paper. We extract various features of the textual contents based on natural language processing technologies firstly, such as word segmentation, keywords, topic model, sentence parsing, distributed word representation etc. Secondly, we build several base-learning classifiers with different features and different machine learning algorithms to assign UGC contents with four different quality labels. Then, we create the global meta-learning model based on these base classifiers to generate the final quality labels for UGC contents. We have also implemented a series of experiments based on realistic data collected from Tianya Forum and use 10-fold cross-validation to test the model. Results have shown that our proposed meta-learning model performs much better.