Intention and Origination: An Inside Look at Large-Scale Bot Queries

Modern attackers increasingly exploit search engines as a vehicle to identify vulnerabilities and to gather information for launching new attacks. In this paper, we perform a large-scale quantitative analysis on bot queries received by the Bing search engine over month-long periods. Our analysis is based on an automated system, called SBotScope, that we develop to dissect large-scale bot queries. Specifically we answer questions of “what are the bot queries searching for?” and “who are submitting these queries?”. Our study shows that 33% of bot queries are searching for vulnerabilities, followed by 11% harvesting user account information. In one of our 16-day datasets, we uncover 8.2 million hosts from botnets and 13,364 hosts from data centers submitting bot queries. To the best of our knowledge, our work is the first large-scale effort toward systematically understanding bot query intentions and the scales of the malicious attacks associated with them.