About the challenge
Dataset Characteristics- Year: Transactions from 2016 to 2018
- Total size: 210,000 transactions
- Additional challenge: Clients who committed fraud in the training set are different from those in the evaluation set
- transactions_train.csv: Transactions for training.
- train_fraud_labels.json: Labels for the training transactions.
- cards_data.csv: Payment card information.
- users_data.csv: User information.
- mcc_codes.json: MCC codes and descriptions.
- evaluation_features.csv: Transactions for evaluation (without labels). (Note: This file is only to be used for model evaluation; it should not be used for training.)
Participants must submit a CSV file containing the following columns:
- transaction_id: Transaction identifier (as in evaluation_features.csv)
- fraud_prediction: Binary prediction (1 for fraud, 0 for non-fraud)
This dataset presents a realistic challenge in bank fraud detection where models must generalize to new clients. Participants will need to develop algorithms capable of detecting fraud for clients never seen during training.
Link to Datas : https://ibm.box.com/s/2oxd8h5vpfghe0iniz6lx06hft5ic12c
Requirements
What to Submit
Submit your source code in a compressed file format (zip).
Ensure that your code is well-documented, including clear instructions on how to run and test your model.
Submission Fomat :
- CSV file with transcation ID , prediction
- Markdown explaining your approach
- notebook with the final work
Evaluation Metrics
Projects will be evaluated based on:
-
Quality of the provided code: Code used for training transactions.
-
Performance metrics: Confusion matrix, Accuracy, Precision, Recall, and F1-Score based on the submission CSV file.
-
Presentation quality: For finalists only.
## Getting Started
### Prerequisites
- **Bob IDE**installed on your laptop ([Download here](https://www.ibm.com/products/bob))
- Basic familiarity with your operating system's terminal/command line
- Git installed for cloning the repository
leaderborad : https://ibm.box.com/s/wa12kd0hra344mno77aommohcbsw1e9g
Prizes
First Team
Second Team
Third Team
Fourth Team
Devpost Achievements
Submitting to this hackathon could earn you:
Judges
IBM
IBM
Judging Criteria
-
Quality of the provided code
Code used for training transactions. -
Performance metrics
Confusion matrix, Accuracy, Precision, Recall, and F1-Score based on the submission CSV file. -
Presentation quality
For finalists only
Questions? Email the hackathon manager
Tell your friends
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.