In this sequel to "An Introduction to Machine Learning with Web Data", bit.ly lead scientist Hilary Mason shows you how to solve real-world problems with machine learning. Using real data from an actual ecommerce website, you will apply production quality algorithms to understand all the issues that arise when working in a live environment.
On Oct 28 Dinesh M wrote: Introduction to Machine Learning!
Disclaimer: I got the free review copy from Oreilly as part of blogger review program.
You cannot get deep understanding of machine learning from these tutorials.
These tutorials are like, What is Machine Learning, What is Machine Learning capable of? This is what you can do with various Machine Learning algorithms.
Machine Learning Experts may get some useful tips. We will get the flavor of Machine Learning starting from third video.
There are totally five videos.The first video is a short introduction to the course. This video is around 19 minutes. Hilary discusses the tools that we will use in this tutorial with the necessary URL to download the source code and the open data set that we will use through out the tutorial.
The second video is around 18 minutes. Hilary introduces terminal along with examples such as how to sort, unique, read and many more examples on the data.
The third video is around 42 minutes. Hilary introduces the concepts of Regression Analysis along with a live example of Decision Trees. There is not much coding here. Hilary used her pre-coded decision tree program to learn and predict about the data.
The fourth video is around 26 minutes. Hilary introduced clustering and what we can do with clustering along with a live example of K-means.
The fifth video is around 21 minutes and covers the concepts of simhashing. Full Review >
On Oct 8 Grant van Staden wrote: An Introduction to Common Machine Learning Practice
The title of this video series promised so much but I'm slightly disappointed.Though many topics are covered, the detail of how each technique works is not conveyed to the extent that you will learn how to implement them. Presentation of the material is a bit awkward but you do get a clear idea of what techniques are available and when to use them. Full Review >
On Sep 22 Jim Schubert wrote: More "Intro to Machine Learning part 2"
If you're looking for an introduction in getting things done with data, you should check out this video. Although the amount of information is pretty light, it is still a good way to get your start conceptually. If you look at the scripts and sample data provided in the code repository, you'll be off to a good start to learn more about your data. Full Review >
On Sep 4 David Witherspoon wrote: Cool ML Algorithms, needs more depth
Advanced Machine Learning video collection is a quick presentation of some machine learning techniques and algorithms covered in just over 2 hours. Even though this is a short period to try and cover any of the many algorithms in machine learning, there is a chance that you might learn something. If you are an experienced in the many topics of machine learning, then you will know that you cannot cover anything in enough detail in 2 hours and therefore this would not be helpful to you. If you are fairly new to the topic, then you might learn a little bit about interesting algorithms, but if you expectation is that you will be able to directly apply them or be able to explain them to co-workers then you will have to dive deeper somewhere else. There were some interesting algorithms that were covered that I had not worked with like Bloom Filter, Simhashing, and Hamming Distance. Hilary explains these algorithms through examples written in Python and utilizing libraries that have implemented these algorithms. The problem is that she does not go into enough detail that you will be able to implement them in another language, therefore you will need to research them to get a better understanding. I did enjoy the advice that she gave about becoming a better data analyst is to watch and talk with other data analyst to see the tools that they use and the approaches that they take. If you take that approach to what she is presenting here, then you will learn some new topics to apply to data mining with the caveat that you will need to spend time researching to better understand the details of the algorithms. Personally I would have enjoyed learning more details about the random forest decision tress and on dimensionality reduction. Since Hilary has the opportunity to create a collection of videos on advanced machine learning, she had the opportunity to dive a bit deeper on the different algorithms and the different situations you can apply them. She could have also taken the time to explain the results that are presented after running the algorithms. Full Review >
On Aug 24 Michal Konrad Owsiak wrote: Very good overview, but too shallow in details
I have seen presentations made by Hilary already (e.g. from Strata) and I think they were very good in terms of being presented as conference materials. During conference, you have limited time and you obviously want to show as much as possible. On the other hand, when it comes to lectures and workshops you have as much time as you can devote to the topic. Full Review >