Introducing the Naive Bayes Algorithm Data Mining

The Naïve Bayes algorithm enables you to quickly create models that provide predictive abilities and also provide a new method of exploring and understanding your data. Thinking about the metaphor from the chapter introduction, it is easy to see how Bayes’ technique can be applied to predictive analysis. Bayes’ paper provides a systematic method for learning based on evidence. The algorithm learns the “evidence” by counting the correlations between the variable you are interested in and all other variables. For example, if you are trying to determine whether a congressperson is Republican or Democrat based on her voting history, your evidence would be the counts of how congress members from each party voted on each issue. The algorithm would then use these counts to form a prediction based on the voting history of the congressperson you were interested in.

Alternatively, you may really be more interested in learning about what issues differentiate the parties. The counts taken by the Naïve Bayes algorithm can be used to explore the relationships among the various attributes in your model. For example, Figure 4.1 shows the top issues that distinguish Democrats from Republicans in the House of Representatives.


Face Book Twitter Google Plus Instagram Youtube Linkedin Myspace Pinterest Soundcloud Wikipedia

All rights reserved © 2018 Wisdom IT Services India Pvt. Ltd DMCA.com Protection Status

Data Mining Topics