{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

hw1_soln

# hw1_soln - CS 478 Machine Learning Homework 1 Suggested...

This preview shows pages 1–4. Sign up to view the full content.

CS 478 Machine Learning: Homework 1 Suggested Solutions 1 kNN Decision Boundaries (a) (b) (c) It would be classified as ‘circle’. Adding one point is enough, say a ‘cross’ at exactly (1,-1). (d) It would be classified as ‘circle’. The five closest points are (0 , 0) , (2 , 0) , (1 , - 3) , (3 , - 2) , (0 , 1). The forth and fifth closest point (3 , - 2) , (0 , 1) are equidistant to (1 , - 1), and of dif- ferent classes. So if we break ties by picking crosses (the point (0,1)), then only one point is needed, say a cross at (1,-1). Otherwise if we break tie by picking the circle then we need at least two points (two crosses at (1,-1), say). 1

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
2 kNN Implementation (a) Always predict the majority class. The baseline accuracy for the Wine dataset is 0.3077, while that for the Liver dataset is 0.6145. (b) See the plots below: The best value of k for the Wine dataset is 7, while that for the Liver dataset is 7 or 13. (c) It might be useful because the input features could have been measured in differ- ent units of varying scales, say, some measured in centimetres while some measured in metres. Assuming those features have similar importance towards classification, the kNN algorithm would give more weight to features with larger range since they contribute more to the Euclidean distance. Rescaling the features with mean and standard deviation could eliminate some of these undesirable effects. 2
(d) See plots below: Standardization improves the performance on the Wine dataset across a whole range of values of k while harms the performance on the Liver dataset.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### What students are saying

• As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

Kiran Temple University Fox School of Business ‘17, Course Hero Intern

• I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

Dana University of Pennsylvania ‘17, Course Hero Intern

• The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

Jill Tulane University ‘16, Course Hero Intern