Exploratory data analysis is an important activity in almost all types of data mining projects. Understanding the distribution, shape of the distribution, and vital parameters of the distribution, is very important. Preliminary hypothesis testing can be done to know the data better. Not only the distribution and its properties but also the relationship between various variables is important. Hence in this chapter, we looked at bivariate and multivariate relationships between various variables and how to interpret the relationship. Classical statistical tests, such as t-test, F-test, z-test and other non-parametric tests are important to test out the hypothesis. The hypothesis testing itself is also important to draw conclusions and insights from the dataset.

In this chapter we discussed various statistical tests, their syntax, interpretations, and situations where we can apply those tests. After performing exploratory data analysis, in the next chapter we are going to look at various data visualization methods to get a 360-degree view of data. Sometimes, a visual story acts as the simplest possible representation of data. In the next chapter, we are going to use some inbuilt datasets with different libraries to create intuitive visualizations.