Hello Everyone,

I am happy to announce first release 0.1.0.Final of Hawkular Data Mining. Release contains several time series models and utility classes for time series modelling:

Time series models

Simple exponential smoothing
Double exponential smoothing (Holt's linear trend)
Seasonal triple exponential smoothing (Holt Winters)
Simple moving average (Weighted moving average)
AutomaticForecaster - which automatically selects the best model
All these models are estimated using maximum likelihood estimation and available for online learning

Time series manipulation & Statistics

Augmented Dickey-Fuller test
Autocorrelation function (ACF)
Time series decomposition
Time series lagging
Time series differencing
Automatic period identification

If you want to try it with Hawkular: https://goo.gl/R93k02

clone https://github.com/hawkular/hawkular/tree/datamining - hawkular/hawkular branch datamining
build with -Pdev && run
enable predictions preferably for every metric https://github.com/hawkular/hawkular-datamining and wait for some time (see below)
go to the Explorer tab, choose metric and increase forecasting horizon

Data Mining is currently not able to query historical metric data from metrics due to authentication issues. However it still receives metrics from bus so it is important to enable predictions as soon as possible and wait for some time to collect data (or increase collection interval). This will be fixed in the next versions.

Thanks to Jiri Kremser and UI team for helping with predictive charts.

Useful links:

Regards

Pavol