Finding Influential Users in Social Networking using Sentiment Analysis

Abstract

Social networking platforms facilitate sharing of information, ideas, and thoughts by constructing virtual communities. Finding people in certain social networking sites, who really have the power to influence other users, is critical. For example, this search can be focused on the right person who can impactfully support or contradict any opinion or generate more profits when they publish an advertisement for a certain business or product. The aim of this study is to devise a computational method of finding the most influential users on social networking platforms. In this study, we propose a solution named Muatheer, which helps determine who influential users are on Instagram. The study uses techniques of scraping data using Instagram Application Programming Interface (API) and applying the sentiment analysis algorithm to classify comments as either positive or negative. The study also uses the unigram as a feature extraction method. Then, the sentiment analysis result will be combined with other factors to calculate the "influence ratio", which attempts to determine the actual influencer in a specific domain. These experiments were conducted using a publicly available datasets from Instagram, and the proposed algorithm delivered a highly accurate result more than 85 %.

Artificial Intelligence-Powered Sentiment Analysis Tool to Assess the Popularity of Political Leaders in Informal Communities

“…Through this analysis, the study seeks to uncover patterns, trends, and underlying emotions embedded within informal political discourse, offering valuable insights into the factors influencing the popularity and perception of political leaders among digital audiences. By exploring the intersection of technology, sentiment analysis, and political communication, this research contributes to our understanding of the evolving dynamics of public opinion in the digital age and informs strategies for political engagement and communication in online environments [3].…”

Recognition of Opinion Leaders in Social Networks Using Text Posts’ Trajectory Scoring and Users’ Comments Sentiment Analysis

“…The present work [21] outlines a novel approach for identifying the most prominent users on Instagram by employing sentiment analysis with a set of criteria. These criteria include the number of likes, number of followers, frequency of keywords, and the positivity of posts.…”

“…[23] Blogs X [24] Forum X The table above 1 presents a comparison between the aforementioned studies. Both articles [21]and [22]focused on analyzing the influence of users in social networks, specifically by considering social signals and comments. However, one crucial aspect that was overlooked in both studies is the analysis of content.…”

A Multi-label Classification of Disaster-Related Tweets with Enhanced Word Embedding Ensemble Convolutional Neural Network Model

“…The work proposed in [15,16] used CNN with word embedding and demonstrated superior performance to hand-crafted features in disaster-related tasks. In the study [17,18], CNN and MLP-CNN with word embedding are used to categorize the data linked to crises. The skip-gram model of the word2vec tool was applied [19,20,21] to extract information from an extensive corpus consisting of almost 57,908 tweets.…”