Determination whether the tasks required are of Supervised or Unsupervised learning and factord helped to take the decision
Used concept of overfitting to explain why when a model is fit to a training data, zero error with those data is not necessarily good
Normalized the data using Numpy, Pandas and Sci-kit learn