Detecting Spammers in Twitter Network

Abstract

The goal of Twitter is to allow friends communicate and stay connected through the exchange of short messages. However, sometimes, spammers also use Twitter as a platform to post malicious links, send unsolicited messages to legitimate users, and hijack trending topics because of two problems of Twitter. These problems are the possibilities to automatically receive following users' updates and to write on followers' profile pages. For this reason, spam is becoming an increasing problem on Twitter day after day as other online social network sites are. In this article, we present several methods to detect spam tweets on Twitter. For this purpose, we utilize Naive Bayes, Random Forest J48, and IBK algorithms. The experiments conducted on real Twitter accounts demonstrate that the Random Forest algorithm gives us the best result to detect spammers in Twitter.

Predicting Rogue Content and Arabic Spammers on Twitter

“…As observed by McCord and Chuah [6], a person can pick the best algorithm to use though empirical examination. Interestingly, Ameen and Kaya [12] carried out a similar exercise and found that random forest had the highest success at 92.95%. A researcher must experiment to determine the best algorithm to use before proceeding with further analysis.…”

“…Digital social platforms suffer from spamming because users with unethical intent use them for self-serving reasons. Commonly targeted sites include "Twitter, Facebook, and LinkedIn" [12]. The main types of spam include the duplication of information, continuous reposts of topics for marketing purposes, and distribution of unwanted messages.…”

Put your money where your mouth is: Using deep learning to identify consumer tribes from word usage

“…Launched in 2006, Twitter rapidly became one of the most popular social media platforms and the dominant microblogging social networking service (Ameen & Kaya, 2017;Case & King, 2011). Over time, it revealed its importance for firms (Hennig-Thurau, Wiertz, & Feldhaus, 2015), which may use it to reach their current and potential customers, advertise their products, and get relevant information for marketing purposes (Burton & Soboleva, 2011;Jansen, et al, 2009;Singh, et al, 2017b;Thomases, 2009).…”