How to Think About Machine Learning Algorithms

If you don't know the question, you probably won't get the answer right. This course is all about asking the right machine learning questions, modeling real-world situations as one of several well understood machine learning problems.

Machine learning is behind some of the coolest technological innovations today, Contrary to popular perception, however, you don't need to be a math genius to successfully apply machine learning. As a data scientist facing any real-world problem, you first need to identify whether machine learning can provide an appropriate solution. In this course, How to Think About Machine Learning Algorithms, you'll learn how to identify those situations. First, you will learn how to determine which of the four basic approaches you'll take to solve the problem: classification, regression, clustering or recommendation. Next, you'll learn how to set up the problem statement, features, and labels. Finally you'll plug in a standard algorithm to solve the problem. At the end of this course, you'll have the skills and knowledge required to recognize an opportunity for a machine learning application and seize it.

• Course Overview 1min
• Introducing Machine Learning 24mins
• Classifying Data into Predefined Categories 28mins
• Solving Classification Problems 31mins
• Predicting Relationships between Variables with Regression 16mins
• Solving Regression Problems 20mins
• Recommending Relevant Products to a User 27mins
• Clustering Large Data Sets into Meaningful Groups 24mins
• Wrapping up and Next Steps 12mins