Machine Learning Summarized in One Picture

#MachineLearning summarized in one picture.

  • Here is a nice summary of traditional machine learning methods, from Mathworks.
  • I also decided to add the following picture below, as it illustrates a method that was very popular 30 years ago but that seems to have been forgotten recently: mixture of Gaussian.
  • Note that you can use a mixture of any distributions, not just Gaussian, for instance, (data-driven) estimated distributions such as those based on kernel density estimation.

Here is a nice summary of traditional machine learning methods, from Mathworks.

I also decided to add the following picture below, as it illustrates a metho…

@bigdataweek: #MachineLearning summarized in one picture.

Here is a nice summary of traditional machine learning methods, from Mathworks.

I also decided to add the following picture below, as it illustrates a method that was very popular 30 years ago but that seems to have been forgotten recently: mixture of Gaussian. In the example below, it is used  to separate the data set into two clusters. Note that you can use a mixture of any distributions, not just Gaussian, for instance, (data-driven) estimated distributions such as those based on kernel density estimation.

Also, I would put neural networks in the supervised learning category.

Top DSC Resources

Follow us on Twitter:  @DataScienceCtrl  |  @AnalyticBridge

A decision tree gives a description of the data. The resulting classification tree can be an input for making a decision. A nice article on decision tree classifiers is this one http://mines.humanoriented.com/classes/2010/fall/csci568/portfolio_… 

I have only one question to that drawing. If decision trees shouldnt be also in classification part? From my point of you they are more often used for classification than regression task.

Machine Learning Summarized in One Picture

You might also like More from author

Comments are closed, but trackbacks and pingbacks are open.