Intro to Machine Learning

Udacity's Intro to Machine Learning is an introduction to data analysis using Python and the sklearn package. The course consists of 15 lessons covering a wide range of machine learning topics including classification algorithms (Naive Bayes, decision trees and SVMs), linear regression, clustering, selecting and transforming features and validation. As a self-paced course, you can take however long you wish on each lesson; some take less than an hour, while others can take several hours depending on how long you work on the mini projects. Intro to Machine Learning requires basic programming and math skills.

Each lesson consists of a series of video segments and quizzes introducing a new topic followed by a mini-project that gives you a chance to work with code implementing the topics you learned in Python using scikit-learn. The course instructors Katie and Sebastian (the guy who runs Udacity) do a good job explaining the material keeping the course engaging, but they keep things simple. The quizzes, at times, are almost patronizingly easy. The mini projects are a bit harder and contribute more to learning, although they occasionally lack adequate guidance and feedback to help students arrive at the expected output. The final project and many of the mini-projects leading up to it, involve detecting persons of interest in the Enron scandal using a data set of emails sent by Enron employees. Interesting real-world data sets are always a plus.

Intro to Machine Learning is an accessible first course in machine learning that prioritizes breadth, high level understanding and practical tools over depth and theory. You won't be an expert in any of the topics covered in this course by the time you're done, but you will be exposed to several major topics in machine learning and have a basic understanding of how they work. If you are interested taking a similar course with many interesting mini projects that uses the R programming language, try MIT's Analytics Edge on edX. Coursera's Machine Learning with Andrew Ng is a logical next step to dig deeper into machine learning algorithm design and implementation, while Caltech's Learning from Data on edX is a great course if you are interested in machine learning theory. Just be aware that both of these courses (particularly the Caltech course) require a stronger math background.

I give this course 4 out of 5 stars: Very Good.

I started this course after having taken the Coursera course of AndrewNg. My goal was to apply the algorithms in Python and to become familiar with Scikit learn. I have completed about 70% of Udacities intro to ML and I have to say I am very disappointed about the quality of the course, especially about the quality of the videos and the quizzes. The mathematical level is broken down to high school level, which is good for the intuitive understanding, but in my opinion the level is far too low to learn anything serious, especially when comparing with AndrewNgs course. The same applies for the quizzes. Let me illustrate this with an example. Assume they want you to calculate a*b/(c*d+e*f). Then there would be a quiz to calculate a*b, another quiz to calculate c*d, another quiz to calculate e*f, another quiz to calculate c*d+e*f, and then finally the whole thing. One has to go through 6 videos and 5 quizzes to calculate a simple fraction. The programming assignments are similar in quality. I have to say I didnt finish the course and therefore I can not comment on the final project, which may be more serious. In conclucion, I can not recommend this course to anyone who has a serious interest in learning something about ML. Invest your time better!!