Saturday, October 12, 2019

Data Science Study Notes: What's the intuitive difference between Bayesian Classifier and Naive Bayesian Classifier?

  • Most people understand the logic behind Bayesian Classifier, it's one of the must-learn Bayes theorem in stats, based on conditional probability. 
  • Naive Bayesian Classifier is a simplified version of Bayesian Classifier, with the big assumption that all features x1, x2, ..., xk are all independent of each other, which is hardly to be true to get a set of predictors which are entirely independent in reality, especially if you didn't understand each feature very well, or didn't do the feature selection first. 




No comments:

Post a Comment

Data Science Study Notes: RFM model to identify the best customers

80% of your sales come from 20% of your customers. As a small business owner, even if you’ve never heard of the Pareto Principle, you know ...