Finding Influencers on Twitter with Using Machine Learning Classification Algorithms
Mehmet Nurettin Şimşek, Abdullah Talha Kabakuş · DergiPark (Istanbul University) · 2018
Microblog sites are environmentswhere people follow people. With this feature, a microblog site is a convenientenvironment for spreading an opinion or introducing a new product. The keypoint is determination of individuals who maximize the spreading. This problem is known as InfluenceMaximization (IM) and has attracted attention of many researchers. Many studiesin the literature have modeled IM problem on graphs for different propagationmodels such as Independent Cascade (IC) and Linear Threshold (LT). However,microblogs like Twitter have their own features. Many works on IM in Twitterderive new metrics from user and tweet features; apply a greedy approach forselecting influencers. In this study, we adopted different approach for IMproblem, and we dealt it as a classification problem. Firstly, we collecteddata on International Women Day 2018;empirically we labeled the users as either influencer candidates ornon-influencers; then we applied classification methods for classifying usersinto one class with using features of users. By this way, we obtained aninfluencer candidates set, which is very smaller than entire dataset. Experimentalresults show that making selection with using same heuristic (namely MF) fromthe reduced influencer candidates set outperforms making selection from entiredataset.