COMMUNITY DETECTION IS ONE OF THE MOST IMPORTANT TASKS IN SOCIAL NETWORKS ANALYSIS. THIS PROBLEM BECOMES MORE CHALLENGING WHEN THE STRUCTURE OF THE NETWORK CHANGES DURING THE TIME. IT IS VERY IMPORTANT TO UPDATE THE STRUCTURES OF THE COMMUNITY IN A DYNAMIC NETWORK WITHOUT TIME-CONSUMING PROCEDURES. THIS PAPER SUGGESTS A HYBRID EVOLUTIONARY ALGORITHM FOR ONLINE COMMUNITY DETECTION. THE PROPOSED ALGORITHM CALLED MEMETIC BASED ONLINE COMMUNITY DETECTION (MBOC) IS BASED ON A MEMETIC ALGORITHM WITH NEW GENETIC OPERATORS AND A NOVEL STOCHASTIC LOCAL SEARCH TO ASSIGN NEW NODES TO COMMUNITIES AND ANOTHER LOCAL SEARCH CALLED DENSE SEARCH TO MODIFY COMMUNITIES AFTER NEW ASSIGNMENTS. THE METHOD IS EVALUATED OVER SEVERAL WELL-KNOWN BENCHMARK NETWORKS. THE RESULTS SHOW THAT THE PROPOSED APPROACH OUTPERFORMS THE PREVIOUS METHODS IN MOST CASES.

We are living in the data age. Communications over scientific networks creates new opportunities for researchers who aim to discover the hidden pattern in these huge repositories. This study utilizes network science to create collaboration network of Iranian Scientific Institutions. A modularity-based approach applied to find network communities. To reach a big picture of science production flow, analysis of the collaboration network is crucial. Our results demonstrated that geographic location closeness and ethnic attributes has important roles in academic collaboration network establishment. Besides, it shows that famous scientific centers in the capital city of Iran, Tehran has strong influence on the production flow of scientific activities. These academic papers are mostly viewed and downloaded from the United State of America, China, India, and Iran. The motivation of this research is that by discovering hidden communities in the network and finding the STRUCTURE of intuitions communications, we can identify each scientific center research potential separately and clear mutual scientific fields. Therefore, an efficient strategic program can be design, develop and test to keep scientific institutions in progress path and navigate their research goals into a straight useful roadmap to identify and fill the unknown gaps.

Nowadays, social networks have gained a lot of popularity among people. With the growth of these networks and a large number of people using these networks, social network analysis has received special attention, so the need for highly accurate and fast algorithms on various issues is strongly felt. One of the important issues in these networks is COMMUNITY DETECTION problem that many algorithms have been proposed for this purpose. In social networks, communities usually are formed around popular or influential nodes. Most algorithms in this field, that are usually density-based, are unable to detect this STRUCTURE. In this paper, we propose a new COMMUNITY DETECTION algorithm based on the local popularity STRUCTURE. In this algorithm, the most popular person in neighborhood of each user is selected as a leader and the user falls into that group. Experimental results on six real networks show that the proposed method not only has comparable results in terms of NMI and ARI, but also has shorter execution time compared to existing algorithms.

NOWADAYS, THE EMERGENCE OF ONLINE SOCIAL NETWORKS HAVE EMPOWERED PEOPLE TO EASILY SHARE INFORMATION AND MEDIA WITH FRIENDS. INTERACTING USERS OF SOCIAL NETWORKS WITH SIMILAR USERS AND THEIR FRIENDS FORM COMMUNITY STRUCTURES OF NETWORKS. UNCOVERING COMMUNITIES OF THE ONLINE USERS IN SOCIAL NETWORKS PLAYS AN IMPORTANT ROLE IN NETWORK ANALYSIS WITH MANY APPLICATIONS SUCH AS FINDING A SET OF EXPERT USERS, FINDING A SET OF USERS WITH COMMON ACTIVITIES, FINDING A SET OF SIMILAR PEOPLE FOR MARKETING GOALS, TO MENTION A FEW. ALTHOUGH, SEVERAL ALGORITHMS FOR DISJOINT COMMUNITY DETECTION HAVE BEEN PRESENTED IN THE LITERATURE, ONLINE USERS SIMULTANEOUSLY INTERACT WITH THEIR FRIENDS HAVING DIFFERENT INTERESTS. ALSO USERS ARE ABLE TO JOIN MORE THAN ONE GROUP AT THE SAME TIME WHICH LEADS TO THE FORMATION OF OVERLAPPING COMMUNITIES. THUS, FINDING OVERLAPPING COMMUNITIES CAN REALIZE A REALISTIC ANALYSIS OF NETWORKS. IN THIS PAPER, WE PROPOSE A FAST ALGORITHM FOR OVERLAPPING COMMUNITY DETECTION. IN THE PROPOSED ALGORITHM, IN THE FIRST PHASE, THE LOUVAIN METHOD IS APPLIED TO THE GIVEN NETWORK AND IN THE SECOND PHASE A BELONGING MATRIX IS UPDATED WHERE AN EACH ELEMENT OF BELONGING MATRIX DETERMINES HOW MUCH A NODE BELONGS TO A COMMUNITY. FINALLY, SOME OF THE FOUND COMMUNITIES ARE MERGED BASED ON THE MODULARITY MEASURE. THE PERFORMANCE OF THE PROPOSED ALGORITHM IS STUDIED THROUGH THE SIMULATION ON THE POPULAR NETWORKS WHICH INDICATES THAT THE PROPOSED ALGORITHM OUTPERFORMS SEVERAL WELL-KNOWN OVERLAPPING COMMUNITY DETECTION ALGORITHMS.

NETWORK CLUSTERING IS ONE OF THE PROBLEMS THAT HAS ATTRACTED MANY RESEARCHERS IN RECENT YEARS. IN THIS ISSUE, EACH USER IS ASSOCIATED WITH A SPECIFIC COMMUNITY BASED ON THE VARIOUS FEATURES OF THE NETWORK, INCLUDING THE STRUCTURE. IN THE RECENT YEARS, DEEP LEARNING IS WIDELY USED TO EXTRACT THE FEATURE VECTOR OF NODES THEN THE VECTORS ARE USED TO FIND THE COMMUNITY OF EACH NODE. IN THIS PAPER, A NETWORK REPRESENTATION LEARNING ALGORITHM IS PRESENTED BASED ON THE INFORMATION OF THE NEIGHBORS OF EACH NODE AND COMMUNITIES ON THE NETWORK. THE RESULTS SHOW THAT OUR NODES’ REPRESENTATION METHOD OFFERS A BETTER QUALITY CLUSTERING OF SOCIAL NETWORKING USERS THAN THE PREVIOUS NETWORK REPRESENTATION LEARNING METHODS.