Community detection in network using chronological gorilla troops optimization algorithm with deep learning based weighted convexity

Community is expressed as the subset of nodes in the graph and the interrelation amongst the nodes within the network is deeper than the remaining of the network. Thus, the detection of community in the network is usually designed as a process for representing the network as a tree. Community refining or detection is very crucial in recognizing and evaluating the model of huge as well as complex networks. In this research, the community detection in the network is done based on the split and merge mechanism with the Chronological gorilla troops optimization algorithm (CrGTO) and deep learning-based weighted complexity. Here, the CrGTO is modelled by adapting the chronological concept in the updated location of gorillas to attain better performance. The Gorilla Troops Optimization (GTO) utilizes the present iteration for updating the location of gorillas. Besides, the Chronological concept utilizes the previous iteration for updating the solution. Hence, the GTO algorithm adapts the Chronological concept for updating the optimal location, such that a better solution is achieved. In addition, the devised model utilizes the weighted convexity as a fitness function where the weight in fitness is determined with Deep Belief Network (DBN). Thus, the invented model acquired better performance based on the MI-based betweenness and computational complexity of 0.679 and 0.540.