Machine Learning: 10 Numerical Recipes

April 8, 2016

The chatter about smart is loud. I cannot hear the mixes on my Creamfields 2014 CD. Mozart, you are a goner.

If you want to cook up some smart algorithms to pick music or drive your autonomous vehicle without crashing into a passenger carrying bus, navigate to “Top 10 Machine Learning Algorithms.”

The write up points out that just like pop music, there is a top 10 list. More important in my opinion is the concomitant observation that smart software may be based on a limited number of procedures. Hey, this stuff is taught in many universities. Go with what you know maybe?

What are the top 10? The write up asserts:

  1. Linear regression
  2. Logistic regression
  3. Linear discriminant analysis
  4. Classification and regression trees
  5. Naive Bayes
  6. K nearest neighbors
  7. Learning vector quantization
  8. Support vector machines
  9. Bagged decision trees and random forest
  10. Boosting and AdaBoost.

The article tosses in a bonus too: Gradient descent.

What is interesting is that there is considerable overlap with the list I developed for my lecture on manipulating content processing using shaped or weaponized text strings. How’s that, Ms. Null?

The point is that when systems use the same basic methods, are those systems sufficiently different? If so, in what ways? How are systems using standard procedures configured? What if those configurations or “settings” are incorrect?

Exciting.

Stephen E Arnold, April 8, 2016

Comments

2 Responses to “Machine Learning: 10 Numerical Recipes”

  1. Dinesh Vadhia on April 9th, 2016 5:34 am

    Worth reading: “Machine Learning: The High Interest Credit Card of Technical Debt ” (http://research.google.com/pubs/pub43146.html).

  2. Donte Zellous on April 23rd, 2016 8:17 am

    Nice post. I find out something more difficult on distinct blogs everyday. It will always be stimulating you just read content using their company writers and use a little from their site. I’d would rather use some while using the content in my blog regardless of whether you don’t mind. Natually I’ll provide you with a link in your web blog. Many thanks sharing.

  • Archives

  • Recent Posts

  • Meta