Two Challenges

This year's contest consists of two classification tasks based on data from anomalous web transactions. In each task, the objective is to rank the examples in the dataset according to the probability each example has a positive label.

You can compete in either or both of these two related tasks.

Click on the tabs above to read about the E-commerce Transaction Anomaly Classification (Easy) and the E-commerce Transaction Anomaly Classification (Hard). The data format(s) are described there and available on the download page.