Search⌘ K
Home
Courses
Business Machine Learning

Data Preprocessing

Explore data preprocessing strategies to improve machine learning models for chronic kidney disease prediction. Understand how to handle missing data using imputation, engineer features through interactions, and create dummies for categorical variables. Gain practical skills to prepare data effectively for predictive modeling in healthcare.

We'll cover the following...

We can do extensive EDA to understand the data well. Let's focus on preprocessing and consider handling the missing data, feature engineering, interactions, creating dummies, and so on. Let's start with converting the target (class column) to 0/1 from notckad/ckd. We can also change the name of the class column to target. However, class is a keyword in Python, and we should not be confused.

Python 3.8
ckd['target']=[1 if i=='ckd' else 0 for i in ckd['class']] # recall list comprehension
print(ckd['target'].value_counts())
# we don't need the class column any more.... lets drop it
if 'class' in ckd.columns:
    ckd.drop('class', axis=1, inplace=True)
    print("Dropping class column.")
else:
    print("The class column is already dropped")

Since we are done with the class column, let’s deal with the missing data values.

Dealing with the missing data

Let’s start with filling the data to create our prototype. We can always refine our models later on for better performances, a continuous ...