I’m a MOOC addict… I try a lot of courses and actually finished them if (and only if!) I feel I’m getting any actionable knowledge. In this page you’ll find courses that I attended and my opinion on them. I will only list courses that I finished and felt they added something to my work in game analytics and/or data science.
This course appears at Lynda.com as an Excel course and it really isn’t. It is about communication and story telling through data visualisation. It is a quick and clean course on how human beings relate with data. Recommended if you create data visualisations.
I love Python, it is my favourite programming language. Unfortunately it is also the language I most often neglect in favour of others that do a better job at whatever I have to do. So I took this course for a refresh. It is an introductory Python course with some extra info on some computer science notions like memory usage and cost and algorithms. The only reason why I completed it was because it was fun (probably the most fun course I had) and the project was to build a search engine which was something new to me. I recommend it if you want to learn Python.
The course has changed since then. Now it includes a social network and the search engine as projects.
A great introductory course by one of the big names in machine learning: Andrew Ng. This course was important to me because my focus at the time I took it was on statistical learning. This course allowed me to understand the different, more algorithmic and less statistical, mindset of “pure” machine learning. If your interest lies mostly on machine learning, this is a great course. It will be less important if you fall more on the side of data analysis. If that is the case you’ll find Statistical Learning (down the page) more relevant.
This is a nice introductory course on machine learning. Think of it as a machine learning 101 with R. It addresses fundamental machine learning building blocks while using the caret package. However I don’t think that this course does enough for someone to be a machine learning practitioner. I’ll recommend it if you don’t have any knowledge of machine learning or if you are interested in the caret package.
A good starting point as an introduction to the R programming language. I recommend it for first steps into R. I’ve tried a number of them and this is the only one that got on this list!
By the time I took this course I was already deep in to R and you should be too! What I took from it was the concept of reproducible research, why and how to do it. This course influenced the way I think about internal peer review in the context of a data science team. It is a very simple course and I’d recommend it if you don’t have an academic research background and feel that reproducibility is important for you as a data analyst/scientist or your team.
This course gave me quite a refresh on statistical inference when I needed it but I can’t say I enjoyed it. I’d love to find another statistical inference course as in depth as this one but with better videos and explanations. The knowledge and passion of the instructor were obvious and that made the course a tad less painful but I almost didn’t take it to the end. I’ll only recommend this course if you want to learn or refresh your statistical inference basics in the context of R.
And if you, dear reader, know of a good course on this, please leave a comment with a link to it!
Statistical Learning (the link changes from time to time, so if it doesn’t work google “Stanford statlearning”)
My favourite course so far. A beast of intuition, math and use cases for supervised and unsupervised learning algorithms. It really doesn’t matter what you know or don’t know or think you know about regression, classification and clustering. This one is a must have for any data analyst/scientist!