Organization, Bot, or Human: Towards an Efficient Twitter User Classification

Abstract

Today, through Twitter, researchers propose approaches for classifying user accounts. However, they have to face confidence challenges owing to the diversity of the types of data propagated throughout Twitter. In addition, the messages from Twitter are imprecise, very short and even written in many dialects and languages. Moreover, the majority of the related works focus on the overall user's activity, which makes them not suitable at the post-level classification. This paper presents an alternative approach for classifying user accounts as being accounts of bots, humans or organizations. The suggested approach consists in accurately classifying user accounts from one single post by leveraging a minimal number of language-independent features. We performed several experiments over a Twitter datasets and supervised learn-based algorithms. Our results demonstrated that simply using a minimal number of language-independent features extracted from one single post is sufficient to classify user accounts accurately and quickly. Our proposed approach yielded high F1-measure (>95%) and high AUC (>99%) using Random Forest.Keywords. Social network analysis, twitter user classification, human vs. bot vs. organization, statisticalbased approach, content-based approach, hybrid-based approach.Human 0,939 0,940 0,939 0.993 Bot 0,986 0,977 0,981 0.995 Organization 0,888 0,919 0,902 0.992

Real-Time Bot Detection from Twitter Using the Twitterbot+ Framework

“…Twitter user accounts have been analyzed across many dimensions. Several works have been done on a single and specific perspective such as those related to the bot detection ( [Varol,17], [Ferrara,17], [Cresci,a17], [Bindu,18], [Kudugunta,18], [Morstatter,16], [Lee,11], [Wu,17], [Jain,19], [Jain,18], [Tavares,17], [Singh,18], [Stukal,17], [Cresci,b17], [Daouadi,b19], [Cresci,15], [Kantepe,17], [Gilani,17], and [Chen,15]) and those related to the effects that bots have ( [Cresci,19], [Mazza,19], [Gilani,19] and [Cresci,18]). The first part of this section discusses the ground truth acquisition methods for bot detection.…”

“…This yielded AUC result of 92.0% using Random Forest algorithm. In [Daouadi,b19], the authors used statistical parameters extracted from the overall user's activity. The best experimental results are obtained with Deep Forest algorithm, this yielded Accuracy result of 97.55%.…”

“…Twitter is one of the most popular social media that allow users to post messages known as tweets. The emergence of Twitter has motivated a large number of research axes such as user account classification [Tavares,13] [Daouadi,a19], topic classification [Khan,17], event detection [Troudi,18] and even terrorist detection [Rekik,18]. This free micro-blogging has been used not only by humans to create and share content, but also by bots (automated agents) to spread misinformation, to pollute content and even to perform terrorist propaganda [Nizzoli,19] [Magdy,16].…”