Improved Dynamic Community Detection based on Distance Dynamics in Real World Network

Computer Engineering

In today’s era, real world networks have become quite prominent and a lot of research is getting done for making these networks valuable, by proposing helpful communities and similar interest communities to users. Moreover, due to freely available web space and interactions, there are different communities per user, making it extremely hard for highly iterative detection algorithms perform quickly and give important suggestions. Furthermore, real world networks are dynamic in nature. Along this lines, there is a requirement for dynamic community detection algorithm which can appropriately detect communities with time differs. Because it accepts the changes in network rather than static community detection algorithm. A Real world network is vital complicated network. There is a number of algorithms developed to detect communities. Dynamic Community Detection based on Distance Dynamics algorithm doesn’t think about different cohesiveness of each neighbor node. In this way, because of the different degree of neighbor node, different attractive strength from the neighbor node to two end points. Here, we propose improved Dynamic Community Detection based on Distance Dynamics (iDC3D) algorithm which use neighbor cohesion for getting different cohesiveness of neighbor nodes. In this, we calculate distances between nodes. Then it uses neighbor cohesion. After that it uses two new interaction patterns for getting the influences between nodes. Then we will get new distances between nodes. We will get communities from the real world network. We demonstrate that the proposed improved Dynamic Community Detection based on Distance Dynamics (iDC3D) algorithm terminated on a decent community number and additionally has comparable detection accuracy with existing approach.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.