-
Book Overview & Buying
-
Table Of Contents
The Data Analysis Workshop
By :
In this section, we perform some preprocessing steps, which will allow us to transform the data into a more human-readable format. Note that data preprocessing and wrangling is one of the most important parts of data analysis. In fact, a lot of hidden patterns and relationships might arise when data is transformed in the correct way.
Furthermore, some machine learning algorithms might not even converge, or they may provide an erroneous result when fed with badly preprocessed data (a typical example of this is not scaling data in deep learning). In other cases, deriving insights from normalized data might be difficult for a human; therefore, it is good practice to transform the data before presenting the results.
In this use case, the data is already normalized and ready for analysis; nevertheless, some of its columns are hard to interpret from a human perspective. Before proceeding further, we will perform some basic transformations on the columns, which will...