Given are the features about each transaction and the elapsed time(seconds) from the first transaction. These features are used to consturct a machine learning model using Logistic Regression.
The given dataset is imbalanced i.e the target class has an uneven distribution of observations(one class label has very high number of observations and the other class label has very low number of observations).
To deal with the imbalanced data, a sample dataset was created using the sample function which randomly selects samples from the original dataset and has equal number of observations as the class with low number of observations.
A prediction system is made and these features are fed to the prediction system and it predicts whether it was a legit transaction or a fraudulent transaction. (0->Legit transaction ; 1-> Fraudulent transaction)