On-Demand Bot Detection and Archival System
| Autor: | Hossein Hamooni, Nikan Chavoshi, Abdullah Mueen |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Service (systems architecture)
Computer science ComputingMilieux_PERSONALCOMPUTING 02 engineering and technology Computer security computer.software_genre World Wide Web Bot herder 020204 information systems On demand 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Social media computer |
| Zdroj: | WWW (Companion Volume) |
| Popis: | Unusually high correlation in activities among users in social media is an indicator of bot behavior. We have developed a system, called DeBot, that identifies such bots in Twitter network. Our system reports and archives thousands of bot accounts every day. DeBot is an unsupervised method capable of detecting bots in a parameter-free fashion. In February 2017, DeBot has collected over 710K unique bots since August 2015. Since we are detecting and archiving Twitter bots on a daily basis, we have the ability to offer two different services based on our bot detection system. The first one is a bot archive API that makes it easy for researchers to query the DeBot's archive. This API can be used to answer various queries: Is a given Twitter account a bot? When was this bot active in the past? Which twitter accounts were detected as bots on a specific date? The second service that we offer is an on-demand bot detection platform which can detect bots that are related to a given topic or geographical location, and report them to the user in few hours. This paper explains all the details of the services we offer on top of the DeBot's bot detection engine. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|