I. Introduction
Twitter bots are accounts on Twitter that are fully or partially controlled by a computer program using the Twitter API. The unique ability of being controlled by a software allows those accounts to generate a large amount of content in a short period of time. Twitter’s terms of service allow legitimate bots to operate on the social media platform as long as they clearly identify themselves as a bot in the user profile section 1. This perk has proven useful to news agencies’ accounts that can produce a high volume of news on a daily basis that is shared with their followers. Malicious bots, which do not identify themselves as bots on their profiles, have utilized this ability to spread fake news [1], [2], spam [3], and offensive content [4] on Twitter. Detecting malicious bots has been a challenging task due to the nature of those bots. They try to evade being automatically detected by turning the software controlling these accounts on and off intermittently and constantly changing their behavior patterns to mimic human behavior [5]. The vast presence of bots on Twitter requires reliable and accurate bot detection methods that differentiate legitimate bots from malicious ones.