What assumptions does naive Bayes make? How would you craft a spam detection algorithm with NB? How would you improve it?

Data Science Interview QuestionsCategory: Data ScienceWhat assumptions does naive Bayes make? How would you craft a spam detection algorithm with NB? How would you improve it?
1 Answers
MockInterview Staff answered 6 years ago
  • Naïve: the features are assumed independent/uncorrelated
  • Assumption not feasible in many cases
  • Improvement: decorrelate features (covariance matrix into identity matrix)

Source

Your Answer

1 + 13 =