Not specified | 16 Weeks
Enrollment Minimum Requirements
- Basic Computer skills
- Access to the internet
DSM012: Supervised Learning - Logistic Regression
In the previous courses we have learnt about data pre-processing, making interpretations from
the data and building supervised learning model using Linear Regression, where the target
variable is continuous in nature. However, in many business problems the target variable is not
continuous in nature and consists of categorical or discrete data types, in such cases we will need
algorithms that can predict categories or classes. Such algorithms that can predict categorical
labels are called classification algorithms.
Supervised Learning Classification course will leverage the power of data with known outcomes to
make models and predict categorical labels for unseen data. This course will focus on training,
assessing and interpreting the outputs of the classification models like Logistic Regression and
Decision Trees, which will enable businesses to make eective data-driven decisions.
After completing this course, you will be able to:
• Understand and appreciate the most widely used machine learning algorithms for
classification problems - Logistic Regression, and Decision Tree.
• Interpret the results of a Logistic Regression model.
• Interpret and visualize the results of a Decision Tree model.
• Assess the model performance of a classification model using dierent performance
• Tune models to improve model performance.
• Participants are expected to have knowledge of basic concepts such as linear regression
• Working knowledge of important Python libraries such as statsmodels, sklearn, pandas,
numpy, matplotlib, and seaborn is required
• Participants are expected to be comfortable with installing Python packages and reading
the Python documentation.
• Logistic Regression
• Introduction to Logistic Regression
• Changing threshold of a Logistic Regression model
• Evaluation of a classification model
• Pros and Cons
• Hand-on Logistic Regression