The primary data sets for IEEE BigData 2019 Cup: Suspicious Network Event Recognition competition are now available! We invite all those interested in cybersecurity to participate in the challenge.
You can upload the solutions until September 29th.
Later this month, we will make available additional data which describe individual log events that correspond to the alerts from the main data table. This data will be quite large (roughly 6*10^9 records), and it will be available on request only to participants who exceed the baseline score on the public leaderboard. We believe that this data will allow you to define even better representations of alerts, and thus facilitate the construction of better prediction models.
I hope you enjoy the competition!
When additional data becomes available?
I can only say that additional data will be released later this month.
We plan to release two more data tables - first, the intermediate one which stores all simple localized alerts generated by SoD's systems, that are associated with the alerts from the training and test data. Then, after some time, we want to release the second set with all individual log events. This set will be the largest one.
I hope that it will be possible to provide the intermediate set at the beginning of next week, but unfortunately, I cannot give any guarantees at this time.