- (Rule-Based Classifier) Consider the decision tree
- Convert the decision tree into a set of classification rule sets
- Are the rules mutually exclusive?
- Is the rule set exhaustive?
- Is ordering needed for this set of rules?
- Do you need a default class for the rule set?
- (Nearest Neighbor Classifier) The table below lists a dataset that was used to create a nearest neighbor model that predicts whether it will be a good day to go surfing
|ID||Wave Size (Ft)||Wave Period (secs)||Wind Speed (mph)||Good Surf|
Assuming that the model uses Euclidean distance to find the k-nearest neighbor, what prediction will the model return for the following query instances when k = 1, 3?
|Wave Size (Ft)||Wave Period (secs)||Wind Speed (mph)||Good Surf|
- (Naïve Bayes Classifier) The table below gives details of symptoms that patients presented and whether they were suffering from
Using a naïve Bayes classifier to determine whether a patient with the above symptoms have meningitis.
- (Bayesian Network) Given the Bayesian Network below, what is the joint probability distribution P(Exercise = Yes, Diet = Healthy, Heart Disease = No, Chest Pain = No, Blood Pressure = low)?
- (Perceptron) Given the perceptron Y = sign(0.4X1+0.3X2+0.7X3-0.5). What is the prediction when X1=0.5, X2=-0.5, X3=0.8?
- (SVM) Given w = (0.4, 0.3, 0.7) and b = -0.5.
- What is the margin of the SVM?
- What is the prediction when X1=-0.5, X2=-0.5, X3=0.8? (2 points)
- (Performance Evaluations) Given the confusion matrix
- Calculate the accuracy
- Calculate the F-measure