Code is available in C++, Python, Ruby, R, and Rust at https://github.com/bhatiasiddharth/MIDAS

Really cool project. I'm doing a real-time version (gets input on stdin, sends output on stdout), will probably release the code soon.

Can this detect DDoS like attacks also?

We handle locality in terms of both source and destination, therefore we should be able to handle both DoS and DDoS attacks.

Looks interesting. Better than some of the static detection algorithms out there

can you list out use cases where we can use this algorithm?

In addition to detecting intrusions, it can detect fake ratings and frauds. Basically finding anomalous and suspicious behavior in any dynamic (time-evolving) graph.

Hi, how will the performance be affected if let's say time ticks are not uniform?

Great question, it will be interesting to try it out. Temporal relations should be affected a bit but MIDAS should be able to detect anomalies.

Are you taking context into consideration? For example, Donald Trump's tweets will be much higher than an ordinary person's.

Yes, we take expected count of a particular user/source node into consideration.