Naive Bayes Classifier

• Classification using conditional probability P(A|B) and Bayes’ theorem

  • P(probability of the presence of class A in the whole) * P(probability of the desired object’s feature being present in class A)
  • Do the above for each class

• The most famous one is the classification of spam emails

• It’s a bit oversimplified, but easy to understand

  • https://qiita.com/ishizakiiii/items/07cc7e463dceb3efe1a1

• Use the occurrence probability of past data to classify

• Linear Model shares many advantages/disadvantages

• It serves as a baseline for large datasets.