Open mre opened 9 years ago
twitter recently announced "Practical and robust anomaly detection in a time series" (see blog post here: https://blog.twitter.com/2015/introducing-practical-and-robust-anomaly-detection-in-a-time-series) The code is on GitHub: https://github.com/twitter/AnomalyDetection It is written in R, so it's not really real-time capable yet. That said, the tool looks quite powerful. Here's a list of things it can detect: https://anomaly.io/anomaly-detection-twitter-r/
Would it make sense to port this algorithm to Go and run it in combination with Morgoth?
twitter recently announced "Practical and robust anomaly detection in a time series" (see blog post here: https://blog.twitter.com/2015/introducing-practical-and-robust-anomaly-detection-in-a-time-series) The code is on GitHub: https://github.com/twitter/AnomalyDetection It is written in R, so it's not really real-time capable yet. That said, the tool looks quite powerful. Here's a list of things it can detect: https://anomaly.io/anomaly-detection-twitter-r/
Would it make sense to port this algorithm to Go and run it in combination with Morgoth?