Virtual Labs

Q1. What is feature extraction in the context of pattern classification?

a: The process of capturing properties of a class that differentiate it from other classes. Explanation

Explanation

b: The process of converting a sequence of numbers into a real-world object. Explanation

Explanation

c: The process of labeling samples during training. Explanation

Explanation

d: The process of storing reference points in the feature space. Explanation

Explanation

Q2. Why are fixed length feature representations more popular?

a: They allow for easy comparison of different samples. Explanation

Explanation

b: They require less computational power. Explanation

Explanation

c: They can handle varying lengths of feature vectors. Explanation

Explanation

d: They adapt better to different problems. Explanation

Explanation

Q3. How does the nearest neighbor classifier make decisions?

a: By comparing a new sample with each reference point and assigning the class of the closest reference point. Explanation

Explanation

b: By comparing a new sample with multiple reference points and assigning the most frequent class label. Explanation

Explanation

c: By weighting the features according to their scales. Explanation

Explanation

d: By transforming feature vectors to variable lengths. Explanation

Explanation

Q4. What is a variant of the nearest neighbor algorithm?

a: k-nearest neighbor algorithm. Explanation

Explanation

b: Weighted nearest neighbor algorithm. Explanation

Explanation

c: Nearest neighbor regression. Explanation

Explanation

d: Dynamic Time Warping. Explanation

Explanation

Q5. Why is it important to scale features to similar ranges in nearest neighbor classification?

a: To ensure that each feature contributes equally to the distance calculation. Explanation

Explanation

b: To reduce the computational complexity. Explanation

Explanation

c: To increase the number of features used in the classification. Explanation

Explanation

d: To avoid transforming feature vectors to variable lengths. Explanation

Explanation

Q6. What is the effect of using binary valued features in the feature space representation?

a: They can make the feature space more sparse. Explanation

Explanation

b: They require more complex distance calculations. Explanation

Explanation

c: They make the feature space more dense. Explanation

Explanation

d: They eliminate the need for scaling features. Explanation

Explanation

Feature Representation