In the previous section, we explored the data using numerical indicators; as a result, we learned several methods of finding important information in the distributions under analysis. Now, we will explore another way to extract knowledge from data through a visual approach. By tracing appropriate charts, from our data, we can identify specific trends even before applying machine learning algorithms. What we will be able to draw from this analysis can serve to focus our study on specific areas rather than performing generic simulations.
Through MATLAB tools, we can explore single-variable distributions using univariate plots such as box plots and histograms, for instance. As well, we can show the relationships between variables using bivariate plots such as grouped scatter plots and bivariate histograms. After plotting, we can customize our plot by adding case names, least squares lines, and reference curves.