Summary: Machine Learning

Read the summary and the most important questions on Machine Learning

• 1 Week 1

• 1.1 Basic Classifiers

  This is a preview. There are 19 more flashcards available for chapter 1.1
  Show more cards here

• Wat voor soort space heeft een classifier?

  Multidimensionale dataspace

• Welke assumptions worden gesteld bij Naive Bayes?

  Independent variables (daarom mag je vermenigvuldigen) 
  
  Gaussian distribution 
  (assumptions kloppen vaak niet = naive) 

• Wat voor dataset kan goed worden onderscheiden met Bayes en QDA?

  Een dataset die omgeven is door de ander

• What is the formula dor precision?

  Precision = TP / TP + FP
  It indicates how many of the predicated positive cases are actually positive 

• What is the F1 score?

  Combined both the precision and sensitivity into a single value, providing a balanced measure of a model's performance --> ESPECIALLY WHEN THERE IS AN UNEVEN CLASS DISTRIBUTION

• What is the formula for the F1 score?

  F1score = 2xTP / (2xTP + FP + FN)

• What is on the x-axis and y-axis of the ROC curve?

  X-axis: TP rate (sensitivity) 
  Y-axis: FP rate (100 - specificity) 

• What is the ideal curve of a ROC curve?

  Hugs the top left corner of the plot indicating a perfect classification model with high sensitivity and high specificity

• What does the steepness of a ROC curve indicate?

  A steeper ROC curve indicates a better model as it shows a higher TP rate for a lower FP rate (confidence intervals are needed)

• 1.2 Imbalanced data

  This is a preview. There are 19 more flashcards available for chapter 1.2
  Show more cards here

• What is the problem if you want to investigate a condition that has a 10% prevalence?

  Even with a stupid classifier; 90% accuracy if we assign all samples to be negative