Fraud credit card data - EDA
This article describes how I used different machine learning models to try to predict credit card fraud and see which works best.
To the reader
The original version of this article was written for Medium.com here
I have attempted to export my articles here. The formatting might be off in a few places
Motivation¶
Well, I am studying data science at AAIC and as a case study, I am required to analyze and write a blog about a problem where I applied Machine Learning models to help solve the problem.
Data Source¶
The data for credit card fraud case study can be found with this link. It is a Kaggle link from where you can download the data and work on it.
Initial Findings (EDA)¶
After importing the necessary packages and reading the data into a pandas dataframe, we start analyzing it.
With the info() method we can see :
- all the columns
- all data types
- whether any columns contain null or not as its values We can also see by the numbers written beside the names that there are no missing data in any of the columns.
Next up we analyse the “time” column in the dataset.
Will update rest of the article soon