Handling Imbalanced Data Classification Problems

49 ratings
Offered By
Coursera Project Network
1,514 already enrolled
In this Guided Project, you will:

Understand the business problem and the dataset to choose best evaluation metric for the problem

Create imbalanced data classification model using SMOTE data resampling technique

Compute to ROC curve and use to adjust probability threshold

Clock2 hours
CloudNo download needed
VideoSplit-screen video
Comment DotsEnglish
LaptopDesktop only

In this 2-hour long project-based course on handling imbalanced data classification problems, you will learn to understand the business problem related we are trying to solve and and understand the dataset. You will also learn how to select best evaluation metric for imbalanced datasets and data resampling techniques like undersampling, oversampling and SMOTE before we use them for model building process. At the end of the course you will understand and learn how to implement ROC curve and adjust probability threshold to improve selected evaluation metric of the model. Note: This course works best for learners who are based in the North America region. We’re currently working on providing the same experience in other regions.

Skills you will develop

  • Predictive Modelling
  • Data Resampling
  • Imbalanced Data
  • Receiver Operating Characteristic (ROC)

Learn step-by-step

In a video that plays in a split-screen with your work area, your instructor will walk you through these steps:

  1. Loading and understanding the dataset

  2. Exploring the dataset

  3. Evaluation metric selection

  4. Creating a baseline model

  5. Resampling techniques for imbalanced datasets

  6. Implementing ROC curve

  7. Adjusting probability threshold

How Guided Projects work

Your workspace is a cloud desktop right in your browser, no download required

In a split-screen video, your instructor guides you step-by-step



View all reviews

Frequently asked questions

Frequently Asked Questions

More questions? Visit the Learner Help Center.