جستجوی مقالات مرتبط با کلیدواژه « link prediction » در نشریات گروه « فناوری اطلاعات »

Link prediction seeks to detect missing links and the ones that may be established in the future given the network structure or node features. Numerous methods have been presented for improving the basic unsupervised neighbourhood-based methods of link prediction. A major issue confronted by all these methods, is that many of the available networks are sparse. This results in high volume of computation, longer processing times, more memory requirements, and more poor results. This research has presented a new, distinct method for link prediction based on community detection in large-scale sparse networks. Here, the communities over the network are first identified, and the link prediction operations are then performed within each obtained community using neighbourhood-based methods. Next, a new method for link prediction has been carried out between the clusters with a specified manner for maximal utilization of the network capacity. Utilized community detection algorithms are Best partition, Link community, Info map and Girvan-Newman, and the datasets used in experiments are Email, HEP, REL, Wikivote, Word and PPI. For evaluation of the proposed method, three measures have been used: precision, computation time and AUC. The results obtained over different datasets demonstrate that extra calculations have been prevented, and precision has been increased. In this method, runtime has also been reduced considerably. Moreover, in many cases Best partition community detection method has good results compared to other community detection algorithms.

Nowadays, online social networks are considered as one of the most important emerging phenomena of human societies. In these networks, prediction of link by relying on the knowledge existing of the interaction between network actors provides an estimation of the probability of creation of a new relationship in future. A wide range of applications can be found for link prediction such as electronic commerce and recommender systems or identification of terroristic relations in social networks. In this article, a new idea is presented for the prediction. It is an integration of the two methods of prediction of similarity score based link and prediction of probabilistic link, which is placed in a new category of link prediction methods. This idea acquires the similarity score between nodes from probabilistic techniques and through using learning automata, and provides better results compared to other criteria methods on standard datasets.

Predicting collaboration between two authors, using their research interests, is one of the important issues that could improve the group researches. One type of social networks is the co-authorship network that is one of the most widely used data sets for studying. As a part of recent improvements of research, far much attention is devoted to the computational analysis of these social networks. The dynamics of these networks makes them challenging to study. Link prediction is one of the main problems in social networks analysis. If we represent a social network with a graph, link prediction means predicting edges that will be created between nodes in the future. The output of link prediction algorithms is using in the various areas such as recommender systems. Also, collaboration prediction between two authors using their research interests is one of the issues that improve group researches. There are few studies on link prediction that use content published by nodes for predicting collaboration between them. In this study, a new link prediction algorithm is developed based on the people interests. By extracting fields that authors have worked on them via analyzing papers published by them, this algorithm predicts their communication in future. The results of tests on SID dataset as co-author dataset show that developed algorithm outperforms all the structure-based link prediction algorithms. Finally, the reasons of algorithm’s efficiency are analyzed and presented.