You are given training data {<1.0,->, <4.0, +>, <5.0,->} over a 1D real-valued input space having two possible output classes labeled + and -. What would a weighted 3NN (3 nearest neighbor) algorithm return as the class for the input query value xq = 3.0? Assume that weight values are given by wi = 1/d2 where d = d(xq,xi) is the usual Euclidean distance metric.
(Solved Homework): You are given training data {, , } over a 1D real-valued input space having two p…
Expert Answer
Don't use plagiarized sources. Get Your Custom Essay on
(Solved Homework): You are given training data {, , } over a 1D real-valued input space having two p…
Get an essay WRITTEN FOR YOU, Plagiarism free, and by an EXPERT!
Given training data: <1.0,->, <4.0,+),<5.0,->
Classes: +, –
K= 3
xq= 3.0
wi = 1/d2
| K-Nearest Neighbor | Class | d=Euclidean distance(xq– Xi) | Weight= 1/d2 |
| 1.0 | – | 2.0 | 0.25 |
| 4.0 | + | 1.0 | 1 |
| 5.0 | – | 2.0 | 0.25 |
class of Xq = -0.25 +1-0.25 => 0.50
Hence class of Xq is +.
