Making data science accessible – Data Munging

By Data Munging we mean the process of taking raw data, understanding it, cleaning it and preparing it for analysis or modelling. It is by no means the glamorous part of data science however if done well it plays a more important role in getting to powerful models and insights than what algorithm you use. So, you've been given a new dataset and are looking to model some behaviors in the data. It is really easy to jump straight in and start running regression or machine learning but this is a mistake. The first step is to really understand the data, starting from a univariate view and slowly building out.