Analysis of the Malicious Bots Market

Abstract

Social media bots can pose a serious threat by manipulating public opinion. Attempts to detect bots on social networks have resulted in bots becoming more sophisticated. A wide variety of types of bots has appeared, which must be taken into account when developing methods for detecting them. In this paper, we present the classification of the types of bots, which we made after analyzing the offers from bot traders in the market. We studied 1657 offers from 7 companies for 5 social networks: VKontakte, Instagram, Telegram, YouTube, and TikTok. Based on this information and descriptions from bot-traders, we are aggregating the types of bots and their key features. We also perform price analysis for different types of bots for the Russian Internet segment. The results show that the main pricing factors are bot action and bot quality. At the same time for different social networks, they affect pricing in different ways. Also, for messengers and social networks in which recommendation algorithms take into account complex actions more strongly, there is a tendency for higher quality bots to perform more complex actions. While in other social networks, the complexity of the action and the quality of bots are not correlating. The results of this study can be useful for developers of tools for detecting bots and determining the cost of an attack.

Keywords

• bot detection
• social network
• market analysis
• bot dataset
• malicious bot