On March 17, we will be launching our IEEE internal coding competition on Kaggle. This competition focuses on binary classification using a bank churn dataset. Here are the details: Link to Competition site: https://www.kaggle.com/competitions/binary-classification-w-bank-churn-dataset-shu/overview Competition Overview: - Objective: Predict whether a customer will continue with their bank account or close it (i.e., churn). For each id in the test set, you must predict the probability for the target variable Exited. - Dataset: The dataset includes various features such as customer credit score, age, tenure, balance, number of products, and more. There is a train.csv file which should be used to train your model, and a test.csv file which should be used to test your model's performance. - Dates: March 17, 2025 - March 31, 2025. All final submissions must be turned in before midnight on March 31. - Submissions: Each participant can make up to 5 submissions per day. There is a submission format that must be followed (please see the sample_submission.csv file on the Kaggle site linked above). Submissions are evaluated on area under the ROC curve between the predicted probability and the observed target. - Prizes: There will be Seton Hall University merchandise (mugs and pens) awarded as prizes to the top 3 scorers in the competition. How to Join: - Click on the link listed above. - Click on the "Join Competition" button on the competition page. - Follow the instructions to accept the rules and join the competition. How to Make Submissions: - Prepare your model and generate predictions based on the provided dataset. This should ultimately generate a CSV file formatted like the sample_submission.csv file on the Kaggle site. - Navigate to the Submissions tab on the competition page and click on "Submit Prediction". - Upload your submission file and click "Submit". We encourage all members to participate and showcase their skills. This is a great opportunity to learn, compete, and have fun! Virtual: https://events.vtools.ieee.org/m/471471