Monday 17:45–18:15 in Main Track

Spammers vs. Data: My everyday fight

Juan De Dios Santos

Audience level:
Intermediate

Description

LOVOO, a dating app, is attractive to spam. These spammers disguise themselves as real users, using believable images and accurate descriptions on their profiles to lure our community into chatting with them. This presentation is on everything Antispam. In it, I will talk about the architecture of the system, data, and the algorithms and methods we employ to fight the spammers.

Abstract

Spam sucks, right? You don't like it, I don't like it, users don't like it, and my boss doesn't want them on the app.

LOVOO, the dating platform I work at, is an attractive target for spammers. These fake users (or bots) disguise themselves as real ones, using believable images and accurate descriptions on their profiles to lure our community into believing they are real people.

To fight them, we have a dedicated group - the Antispam team - whose purpose is to research and develop techniques that rely heavily on machine learning and massive real-time data processing to avoid the proliferation of spammers in the platform.

This presentation is on everything Antispam. In it, I will talk about the Kafka and Google Cloud-based architecture of the system, data, our algorithms and methods we employ to kick spammers, and lastly, I will show some insight and numbers about our performance.

Subscribe to Receive PyData Updates

Tickets

Get Now