Most of the problems that data scientists must solve involve how much or how many? (regression), which category? (classification), what’s wrong with the data? (anomaly detection), and would a user prefer this? (recommender systems). The first step to solving any of these problems is to determine what type of data is at hand. So, if I am asked to compare the number of companies with their associated revenues, which graph would I use?
When I first read about the data science course offered by Flatiron school, I immediately thought about the research process that takes place in universities: devising an experiment, collecting data, analyzing the data, organizing it, drawing a conclusion from it, and presenting it. I had experience in the process, so I knew I would enjoy it. Although, I will admit that I thought it only applied in this field alone.