Super Data Science: ML & AI Podcast with Jon Krohn avatar

Dexa/Super Data Science: ML & AI Podcast with Jon Krohn

Learn more
Published Sep 10, 2021

SDS 504: Classification vs Regression — with Jon Krohn

Jon Krohn delves into the realms of machine learning, distinguishing between classification and regression in supervised learning. With practical examples, he elucidates how models predict outcomes, whether categorizing sentiments or recognizing digits, offering insights into the roles of these techniques in data prediction.
Episode Highlights
Super Data Science: ML & AI Podcast with Jon Krohn logo

Topics covered

Popular Clips

Episode Highlights

  • Binary Classification

    introduces the concept of binary classification, a fundamental category in supervised learning. He explains that binary classification involves predicting one of two possible outcomes, such as determining whether a movie review is positive or negative. This type of problem is essential for understanding how models use input data to predict specific classes 1.

    Binary classification, two classes, like movie review sentiment, multi class problem, more than two classes.

    ---

    Binary classification problems can be coded with integers, where outcomes like positive or negative reviews are represented as 1 or 0, respectively 2.

       

    Multiclass Classification

    Multiclass classification extends the binary framework to scenarios with more than two possible outcomes. uses the example of handwritten digit recognition, where the task is to identify digits from 0 to 9, making it a multiclass problem 1. This approach requires models to handle multiple classes, each representing a different category or label.

    So, an example here would be if we had a bunch of handwritten digits.

    ---

    In these problems, the model's goal is to accurately classify inputs into one of several categories, demonstrating the versatility and complexity of supervised learning 3.

Related Episodes