Search

With the rapid development of information technology, scientific social network, i.e. SSN, have become the fastest and most convenient way for researchers to communicate with each other. However, a lot of published articles are shared via SSN every day which make it difficult for researchers to find highly valuable articles. To solve above information overload problem, a novel hybrid approach integrating with social information, i.e., HAR-SI, is proposed for the article recommendation in SSN. Unlike the traditional CBF and CF recommendation approaches, the social tag information and social friend information, which have been proved playing a significant role on recommendation in many domains, are effectively integrated into these two approaches separately to improve the accuracy in SSN. Prediction results made by the improved CBF and CF separately are combined with a hybrid method. In order to verify the effectiveness of the proposed HAR-SI, a real-life dataset from CiteULike was employed. Experimental results show that the proposed approach provides higher quality recommendations than the baseline methods, thus providing a more effective manner to recommend articles in SSN. [ABSTRACT FROM AUTHOR]

With the rapid development of information technology, scientific social networks (SSNs) have become the fastest and most convenient way for researchers to communicate with each other. Many published papers are shared via SSNs every day, resulting in the problem of information overload. How to appropriately recommend personalized and highly valuable papers for researchers is becoming more urgent. However, when recommending papers in SSNs, only a small amount of positive instances are available, leaving a vast amount of unlabelled data, in which negative instances and potential unseen positive instances are mixed together, which naturally belongs to One-Class Collaborative Filtering (OCCF) problem. Therefore, considering the extreme data imbalance and data sparsity of this OCCF problem, a hybrid approach of Social and Content aware One-class Recommendation of Papers in SSNs, termed SCORP, is proposed in this study. Unlike previous approaches recommended to address the OCCF problem, social information, which has been proved playing a significant role in performing recommendations in many domains, is applied in both the profiling of content-based filtering and the collaborative filtering to achieve superior recommendations. To verify the effectiveness of the proposed SCORP approach, a real-life dataset from CiteULike was employed. The experimental results demonstrate that the proposed approach is superior to all of the compared approaches, thus providing a more effective method for recommending papers in SSNs. [ABSTRACT FROM AUTHOR]