It used to be that crawling social media required some creative coding, a deep understanding of links and relationships, and a finely tuned BS detector. With machine learning becoming common place, it turns out you only need the first two.

Considering information in isolation never gives us a complete picture. As we supplement what we know with what we discover with what inferences and patterns our models can detect, what we know and what we KNOW becomes starkly different.

In this session, we'll dive into a multi-year project that has used social media and other public data sources to uncover hiring patterns, stealth startups, and even pending legal action.


Comments are closed.

Josh Butts at 14:43 on 21 Apr 2017

Classic Keith Casey.

Really cool talk. Had me at the two pages of legal disclaimers.

Keith has an exceptional skill to scare audiences in a friendly, smiling way. Superb talk!

Joe Ferguson at 14:31 on 22 Apr 2017

Great talk that covers complex topics that were explained very well.

Huy Nguyen at 15:26 on 22 Apr 2017

I always knew the Internet was evil and only gave the illusion of privacy, but Keith makes our (ethical/legal) invasion of privacy look like fun now. (Love them disclaimers) It's a good thing he is on our side, I think. I now feel empowered to aggregate big data myself and (ethically/legally) crush my enemies and hear the lament on their social media.

Bob Phillips at 14:24 on 25 Apr 2017

I cried.

I'm really upset this was the first time I heard Keith talk. Seriously, he's fantastic, and this talk will make you want to both start analyzing social media patterns and drop off the grid.