Machine Learning India is organizing an AI Challenge for all the beginners in data science. This challenge provides an opportunity for all the data scientists to learn, compete, and showcase their skills at the national level. The dataset consists of Credit Card Transactions that need to be classified into fraudulent and non-fraudulent transactions. This challenge has applications in the domain of risk analysis, insurance claims, loan approvals, and other economic fronts.
The provided dataset has the following 3 files:
TRAIN.csv (contains training data) TEST.csv (test data for prediction) sample_submission.csv (an example of submission file) The TRAIN.csv has 31 attributes, 30 training, and 1 target attribute (Class). "0" represents non-fraudulent transactions and "1" represents fraudulent transactions. Attributes V1...V28 are anonymous so that the internal details may be kept hidden. Competition link(for data set): https://dockship.io/challenges/5f8817fc2ff55918a0995333/machine-learning-india-ai-challenge/overview Or Kaggle link for data set: https://www.kaggle.com/mlg-ulb/creditcardfraud