Quad Motif-Based Influence Analyse of Posts in Instagram

Abstract

Word of epidemic is not specified for real life; it can be seen in virtual networks. Social networks are very complicated networks and extracting information is very important for analyze human behaviors. Finding user's influence on other users is always interesting area in social networks analyses. Almost all previous works focus on calculating user profiles. However, all posts have different influence values. We propose a method finding most effective posts which based on fuse motif analysis (FMA) for The Instagram Social Network. We focus to measure which post have more effect on other people. In our study, we take into account variety of factors having effects on this process in this method. These factors include users' other posts popularity, tag frequency and emotional based sentimental analyze on comments. We try to create a unique model to predict most influencing posts with all these factors. We collect a huge amount of data from the Instagram and share our experimental results on this data.