Numenta Anomaly Benchmark Competition

[Alexander Lavin]

Hello all,

The Numenta Anomaly Benchmark (NAB) is an open-source dataset and scoring methodology designed for evaluating anomaly detection algorithms for real-world use. We’re hosting a NAB competition in conjunction with the IEEE World Congress on Computational Intelligence: numenta.org/nab. We’re offering cash prizes for two categories:

- Algorithms for detecting anomalies in streaming data

- Real-world data sets labeled for anomalies

Entries are due by July 1st. Send submissions and questions/comments to n...@numenta.org; send them early so we can help you! See the competition page (linked above) for more details. You can learn more about NAB and the competition in this short video: youtube.com/watch?v=b0GS_Fbsvzw.

We designed NAB to be useful for the research community; everything is open-source and it’s easy to run your anomaly detection algorithms on NAB (in any language). Check out the repo: github.com/numenta/NAB.

Cheers,

Alex

--

Alexander Lavin

Software Engineer

Numenta