Hacker News Reader

For those of us not familiar with ML, what are some examples of data that is best consumed by something like this?

I could imagine piping system stats into this and then setting up alerts when something "unusual" is happening.

The super amazing thing would be multivariate analysis of multiple series so you could see when the relationships between variables becomes anomalous as well.

We have just that type of model coming up in our next release. So for example in the case of a smart car, a higher value for torque than usual might be OK if it is simultaneous with a peak in the gas pedal. But if a peak in the pedal does not coincide with a peak in torque, that is a problem. Follow us to stay up to date!

Reply on news.ycombinator.com