theashbhat

For those curious like me, here are the Wikipedia entries for David Mayer

> David Mayer may refer to: (A) David Mayer (historian) (1928–2023), American-British theatre historian (B) David Mayer de Rothschild (born 1978), British adventurer, ecologist, and environmentalist (C) David R. Mayer (born 1967), American politician Akhmed Chatayev (D) David Mayer (1980–2017), Chechen Islamist and terrorist

source: https://en.wikipedia.org/wiki/David_Mayer

hyyype

The curation done for the campaign is quite incredible

One of the creators here: Yup! Our current users are more engaged and active on Twitter than the average users.

The chrome extension and website release also helps improve our model significantly. We get feedback on how it works in the wild and false positives that we can use to improve our model on.

Hey! Creator here.

We agree - as independent of Twitter we have a considerable amount of freedom in how we build this. However Twitter can and should be doing much more. For example with a model they can start placing captchas before Tweets.

In addition to building the model we went about trying build our own bots. To do this we went on forums and contacted individuals selling "aged" accounts. It turns out its as simple as sending $4 over paypal to get a compromised account. These are accounts with histories, real followers, and real people behind them.

We bought 11 of these and were able to automate them within the hour of purchasing them. They also started receiving replies to their retweets and content almost immediately from all over twitter.

The ease of setting up these compromised accounts as bots was also incredibly worrisome. We've found high confidence heuristics to determine that an account has been compromised. If we can - Twitter should be able to as well.

We're ultimately a bit confused over Twitter's inactivity here. We also haven't heard anything from the company.

Here's our analysis if you want to read more.

analysis: https://medium.com/@robhat/an-analysis-of-propaganda-bots-on...

Hey! One of the creators here!

Your right - twitter could do (and probably are doing) everything we're doing. They have billions of dollars and hundreds of engineers.

The value of building a model is that we can do a wide analysis on bot like activity. Separately launching botcheck.me as something that users can use is incredibly valuable from the ML side. Users essentially hand classify a bunch of false positives for us (to further train on) and also give us an idea of how are model is doing.

We aren't just doing sentiment analysis and you're right - NLP is hard. Fortunately at UC Berkeley we have some amazing CS professors that have been incredibly helpful in advising us while building this.

We're using LSTMs to learn the weights of various words. We've been using high confidence heuristics to generate our training data that aren't based primarily on tweet content.

One such example is looking at compromised accounts that have had their usernames changed.

Here's an analysis: https://medium.com/@robhat/an-analysis-of-propaganda-bots-on...

Methodology: https://medium.com/@robhat/identifying-propaganda-bots-on-tw...

We want to release a portion of training data so that others can build similar services. Let me know if you have any more questions :)

Creator here! We do identify bot-like activity on both sides of the spectrum (pro-hillary and pro-trump).

hey! One of said creators here! We haven't figured out how to monetize / properly fund this yet. It's more so a problem we saw and attempted to solve. The cost right now is a couple hundred dollars month in server costs.