International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 10 Issue: 07 | Jul 2023
www.irjet.net
p-ISSN: 2395-0072
New Similarity Index for Finding Followers in Leaders Based Community Detection Sunil Patel1, Dr Kumar Gaurav2 1M.tech Scholar, Dept. Of Electronics Engineering, HBTU, Uttar Pradesh, India 2Assistant Professor, Dept. Of Electronics Engineering, HBTU, Uttar Pradesh, India ------------------------------------------------------------------------***------------------------------------------------------------------------Currently the goal of banks and business houses is to Abstract - Currently, the problem of social research has
emerged as a challenge, there have been various debate to understand and analyze human relationships in the network. Many researchers have found ways to search the community and select a leader individually, but the algorithm discussed in this paper is suitable for detection of community and community leaders. Generally speaking, people love to connect with people who share the same behaviour of interest, and to form communities based on shared ideas. A small number of residents in the community are in charge of spreading consciousness there, i.e. leader, they represent a group who have conveyed their thoughts and experiences. The existing algorithms are based on modularity optimization techniques. Leader based community detection method is not modularity optimization dependent techniques. The accuracy of this technique decreases as network size increases so this paper tells about a modified similarity indexing in leader based selection method to increase the accuracy of the method. KeywordsโCommunity Method, LFR Benchmark
Detection,
Similarity
1. INTRODUCTION A network is defined as group or connection system between people or things. These include a number of devices such as computers, servers, peripherals, humans, etc. These things are known as nodes in the context of network science, and the connections between them are known as edges. The internet is one of the best examples of a network since it connects millions of individuals to a single platform and offers a variety of information and services, including the World Wide Web and e-mail. The people gather together in a place called social network. People tend to connect with people who share the same personality or interest. They form a group of people who have common interest and interact more with each other as compared to others, these group of people are called community. When people within a community interact more, this is referred to as an intercommunity link, and when those outside the community interact less, this is referred to as an intercommunity link. ยฉ 2023, IRJET
|
Impact Factor value: 8.226
|
look for active groups in their network as well as in their customer network. In many communities, some nodes play an important role in innovation, with knowledge and ideas shared in the community. These nodes act as a catalyst to cause turmoil in community. Many researchers look for this catalyst node or most important person in the community [1].
Identifying useful nodes in the network is one of the biggest tasks. In biological system, identification of important node in communities is important, for example in cancer therapy, which require the identification and destruction of cancer cells from the blood and require the maintenance of the body. The second example is the September 11 attack, which involved a network of 62 nodes and 153 connections made up of 5 communities. To keep these communities in control the owners of these communities should be kept in check. Finding the most influential node in a network depends on the network's structure and centrality. In order to determine a node's power within a network, centrality is utilized to determine which node is the most influential. There exist two types of centrality local centrality and global centrality [2]. Local centrality includes degree centrality and betweenness centrality. The total number of focal nodes connecting one node to another is used to determine degree centrality. The shortest path between each pair of nodes is used to determine betweenness centrality. Closeness centrality is a sort of global centrality that has an inverse connection with the sum of shortest distances between network nodes. This influential node acts as tools in community to share ideas and information making community powerful.
2. LITERATURE REVIEW The research related to community detection has been started by Girvan and Newman [3]. They proposed GN algorithm in 2002 that detects the community by continuously removing the edges according to the betweenness centrality but this algorithm cannot be applied to large network. They also coined the term "modularity" to describe the strength of a community's ISO 9001:2008 Certified Journal
|
Page
1107