Xente Fraud Detection Challenge
$4,500 USD
Accurately classify the fraudulent transactions from Xente's e-commerce platform
1203 data scientists enrolled, 547 on the leaderboard
Financial ServicesClassificationStructured
Uganda
20 May 2019—23 September 2019

Xente is an e-commerce and financial service app serving 10,000+ customers in Uganda.

This dataset includes a sample of approximately 140,000 transactions that occurred between 15 November 2018 and 15 March 2019.

One of the challenges of fraud detection problems is that the data is highly imbalanced. See these blogs for examples on how imbalanced data might be handled:

The files for download here are:

  • Xente_variable_definitions.csv: Definition of the features per transaction
  • Training.csv: Transactions from 15 November 2018 to 13 February 2019, including whether or not each transaction is fraudulent. You will use this file to train your model.
  • Test.csv: Transactions from 13 February 2019 to 14 March 2019, not including whether or not each transaction is fraudulent. You will test your model on this file.
  • sample_submission.csv: is an example of what your submission file should look like. The order of the rows does not matter, but the names of the TransactionId must be correct. The value in FraudResult will be 1 for is a Fraud and 0 for is not a fraud.