Consider using kNN density estimation to separate two equi-likely classes using l = 1 feature. You...

   

Transcribed Image Text:

Consider using kNN density estimation to separate two equi-likely classes using l = 1 feature. You are given four training values, two for each class: X (1) [-2 1] X(²) = [35]. = Use k = 2 to classify the test value x = 1.6. Specifically a. On the real line, plot the two training sets (e.g., x's for class 1, o's for class 2). Add the test value. b. Compute p(x|w₁). c. Compute p(x|w₂). d. Use these results to determine the class of the test value. Hint: For 1-dimensional feature vectors, "volume" is determined by the length of a line segment centered at the test value. Consider using kNN density estimation to separate two equi-likely classes using l = 1 feature. You are given four training values, two for each class: X (1) [-2 1] X(²) = [35]. = Use k = 2 to classify the test value x = 1.6. Specifically a. On the real line, plot the two training sets (e.g., x's for class 1, o's for class 2). Add the test value. b. Compute p(x|w₁). c. Compute p(x|w₂). d. Use these results to determine the class of the test value. Hint: For 1-dimensional feature vectors, "volume" is determined by the length of a line segment centered at the test value.

Answer rating: 100% (QA)

a On the real line plot the two training sets eg xs for class 1 os for class 2 Add ... View the full answer