No abstract

Recently, social networks have received dramatic interest. The speed of the development and expansion of the Internet has created a new topic of research called social networks or online virtual communities on the Internet. Today, social networking sites such as Facebook, Twitter, Instagram and so forth are dramatically used by many people. Since people publish a lot of information about themselves on these networks, this information may be attacked by the intruders, so the need of preserving privacy is necessary on these networks. One of the approaches for preserving privacy is the K-anonymity. Anonymization always faces the challenge of data lost, therefore, an approach is required for anonymization of data and meanwhile maintaining the usefulness of the data. In this research, by combining the k-anonymity priority clustering method and Cuckoo optimization algorithm, an appropriate model is developed to maintain the privacy of the data and its usefulness. The average path length, average clustering coefficient and the transitivity criteria have been used to evaluate the proposed algorithm. The results of the experiments show that the proposed method in most cases has 1 unit superiority in terms of k-anonymity and 2 units superiority in terms of usefulness in comparison with similar methods.

Social networks play an important role in human life, and the study of communities within them is of particular importance. In this research, sustainable communities called popular communities have been introduced and studied. In this regard, the life of social networks was considered in time snapshots. Then, the communities that which at least 45% of their members were present the next time snapshot were considered popular communities. In the next step, the structure of popular communities and their distinguishing features from other communities as well as some of their structural features along with the other centrality features were introduced. Finally, using the theory of rough set, the importance of the structural features was examined to predict popular communities using the rules derived from this theory. Also, the importance of popular nodes introduced in this study in terms of, Integrated Value of Influence (IVI), Collective Influence (CF) and Spreading Score was compared with the maximum nodes of these values in each of the popular communities. In this study, we tested the performance of our proposed model on the actual Facebook database. We used rough set theory to generate the rules and model learning. Leader-Node, Popular-Node, Closeness, Eigen Centralization, Betweenness and Community Density were used as features and Popular Community was used as a label. In the experiment, 10-Fold Cross-Validation was used for model learning and Standard Voting was used as a classifier in the ROSETTA toolkit environment. The experiment has shown that if the average of an entire community is low on closeness, betweenness, and eigenvector centrality but with a high number of popular nodes, the chances of the community remaining popular increase. In addition, the high density of the community is not a good criterion for the community to remain popular at all. It was also shown that with the help of these features, the popular communities were predicted with almost 84% accuracy. On the other hand, the nodes with the maximum values of IVI, CF and Spreading Score were a subset of popular nodes. This shows the importance of popular nodes in social networks which introduced in this study.

A dynamic Online Social Network is a special type of evolving complex network in which changes occur over time. The structure of a community may change over time due to the relationship changes between its members or with other communities. This is known as a community event. In this paper, we discussed the effect of important individual community features and the lengths of adequate time intervals considered in the analysis of the behavior of social networks on the prediction accuracy of each event. Furthermore, we introduced the extra-structural features as global social network features to justify the relationship between the lengths of time intervals used in the model training by using the best prediction accuracy of events. We found a relationship between the scale of network dynamics and the length of time intervals for observing the spread and decomposed events. Finally, by comparing the accuracy of the model based on time interval length which investigated based on cps-value in this study and using the Event Prediction in Dynamic Social Network (EPDSN) model, the hypothesis of a reverse relationship between cps growth rate and time interval length to obtain better prediction accuracy for both the spread and decomposed events.