Analysis of Social Media Complex System using Community Detection Algorithms

Abstract

The increasing significance of social networks has led to information propagation and community formation being an interesting domain in data science. The data gathered from big social networks exhibit different community structures. These communities attract various users who grow complex networks. The main goal is to identify the impacting nodes responsible for community data flow. The Twitter network edges are considered in the study, which plays a vital role in representing the activities and relationships developed among the community members. Different communities evolve when the network is analyzed using different community detection algorithms. The network statistics are used for analysis by calculating the weighted degree distribution of nodes in this study. The network is analyzed according to persistent clusters using community detection algorithms like the Spinglass, Walktrap, Fastgreedy, Leading Eigenvector, Multilevel, Edge Betweenness, and Label Propagation. It is found that these measures are very useful in community detection and observing the spread of information in social networks.

Barriers to Enduring Pro-Environmental Habits among Urban Residents

“…Based on sociological ideas, some researchers looked at how people behave in favor of the environment, highlighting the importance of social contact, social capital, and other elements [61]. This is very important to find the influential nodes in any big network, such as social me-dia networks to propagate the information in a useful way [62,63]. As a result, social interaction theory has emerged as a critical theoretical framework for understanding how social context shapes pro-environmental behavior.…”

Machine learning techniques for identification of carcinogenic mutations, which cause breast adenocarcinoma

“…In a decision tree root, nodes can be used as input. These nodes are filtered through decision nodes and leaf nodes used for getting desired output 35 – 37 . Entropy controls how data will be split in the decision tree, and information gain tells how much information a feature gives about the respective class.…”