ActiveFence

A fascinating discussion today at PyData Global with ActiveFence VP Data Matar Haller & Big Data Engineer Noam Levy in their presentation on constructing and querying a data model to detect . They discussed how the technology behind detecting online is multi-layered, as is content that users generate. A typical social post has text, an image, and interactions; each must be assessed against algorithms by the model to define a risk score that ranks harmful content.

This data model supports teams scaling their efforts to catch malicious content by calculating the probability of risk.

Online child predators communicate in a variety of unique codes. How can trust and safety teams protect young users?

Read ActiveFence's latest blog by the Head of the Child Safety and Human Exploitation vertical Avi Jager, “How are Pedophiles Hiding in Plain Sight on Platforms?”

It focuses on what teams are doing to identify secret codes used by pedophiles and the importance of on platforms to effectively detect and remove violative content.

How Are Pedophiles Hiding In Plain Sight on Platforms? - By using codewords and esoteric language, child predators have found new ways to go undetected on platforms.