Finding Influential Users in Social Networking using Sentiment Analysis

Abstract

Text of the abstract:Social Networking platforms facilitate the sharing of information, ideas, and thoughts through constructing the virtual communities. Therefore, finding people in given social networking who are really have the power to influence others is more critical. For example, finding the right person who has an impact to support or contradict any opinion or bring more profits when he/she publishes an advertisement for a certain business or product. This problem can be modelled as finding influential users in social networking. There are different ways to find the influential users in these platforms and there are some criteria that should be considered. In this article, we propose a solution named Muatheer which help to determine the influential users in Instagram by scraping data using Instagram API and applying the sentiment analysis algorithm to classify the comments whether it is positive or negative, as well as using the unigram as a feature extraction method. Then, the sentiment analysis result combines with other factors to calculate the “Influence Ratio” that proposes to determine the actual influencer in a specific domain. The experiments were conducted using set of training datasets and the proposed algorithm gives a high accurate result using some metrices.