Introduction
Welcome to this session on ‘Multiclass Classification’. In the previous sessions, you learnt what logistic regression is. You also learnt how to implement logistic regression in univariate and multivariate models using a sigmoid curve. In all those sessions, you have seen the cases that had two target classes only. A sigmoid curve could explain classes such as diabetic or non-diabetic in the case of diabetes example, and churn or not churn in the case of telecom churn example.
In this session, you will learn about classification models that involve multiple, i.e., greater than two target classes. Let us hear from our industry expert Ankit Agarwal, as he introduces himself, explains multiclass classification, and provides a mindmap of the session.
So, as Ankit mentioned, in this session, you will study the concepts of multiclass classification and methods to apply them. We will discuss the following topics in chronological order:
- Practical situations where we need multiclass classification
- Comparison with binary classification and how to derive it using binary models
- Different algorithms used for multi-classification
- Python implementation of these algorithms with a case study
Prerequisites
In this session, we assume that you are familiar with the concepts covered in the previous sessions on logistic regression.
Guidelines for in-module questions
The in-video and in-content questions for this module are not graded.