Identifying Influential User in Twitter : Analysis of Tweet Content Similarity in Weighted Network

In recent years, Social Network Analysis (SNA) is still growing rapidly. The mapping and measurement of the interaction in SNA can be used in many areas, for example to find the most influential users to improve the marketing strategy in Small and Medium Enterprise (SME). In order to find the most influential users in a network, we can apply the centrality measurement such as degree centrality, betweeness centrality, closeness centrality and eigenvector centrality. In this manner, degree centrality is conceptually the simplest one, which is defined as the number of links incident upon a node. While recent works has focused on number of nodes with the weighting between nodes according to its interaction such as following, followed, mention, retweet and reply. In this study, we investigate the combination of tweet content similarity and the interactions between users in twitter using Opsahl method. In this paper, we compare the proposed method with the baseline system from previous research. The experimental result show that the tweet content similarity affect the result of the most influential user in comparison with existing method.