Hey Matt,
cool competition! Got a question, can a synthetic benchmark (the repo we've
been working on..) enter as a dataset? It meets all the criteria:
We will evaluate your dataset against the following characteristics:
- meeting the requirements described above
- interesting types of anomalies
- relevance to real applications
- accuracy/quality of anomaly labels
- good challenges for detection algorithms
- quantity of data — the more files in your submission the better
- feasibility for algorithms to detect anomalies
- We can not expect noisy data without temporal patterns to be
valuable in evaluating algorithms, and thus this type of data will not be
useful in NAB.
But it's not a real-world data?
Cheers,
Mark
On Mon, Feb 15, 2016 at 11:56 PM, Matthew Taylor <[email protected]> wrote:
> Hello NuPIC,
>
> Hopefully, you've heard of the Numenta Anomaly Benchmark (NAB). It is
> an anomaly detection benchmark we here at Numenta created to try to
> compare NuPIC with other temporal streaming anomaly detection
> algorithms. We could not find any benchmarks for streaming temporal
> data, so we created our own!
>
> But now we want to try to get MORE DATA and MORE ALGORITHMS into the
> Numenta Anomaly Benchmark. That is what the competition is all about.
> For all the details, see http://numenta.org/nab/.
>
> You could win up to $5,000 USD if you win both the dataset and
> algorithm category! Please have a look at the contest and share via
> social media to spread the word. The more data and algorithms we have
> available in NAB, the better we all get at machine intelligence. This
> is a chance for you to help out and win money!
>
> http://numenta.org/nab/
>
> Regards,
> ---------
> Matt Taylor
> OS Community Flag-Bearer
> Numenta
>
>
--
Marek Otahal :o)
