Exploring KNN and SVM Concepts

What does KNN stand for in machine learning?

• K-Nearest Neighbors Algorithm

• K-Nearest Neighbors

• K-Nearest Network

• K-Nearest Nodes

Explain the basic principle of the KNN algorithm.

• KNN uses a single nearest neighbor to classify data points.

• KNN requires labeled data to be sorted before classification.

• KNN classifies data points based on the majority class of their k nearest neighbors.

• KNN predicts the value of a data point based on its distance from the origin.

How does the choice of kernel function affect SVM performance?

• The kernel function has no effect on SVM performance.

• All kernel functions perform equally well in SVM.

• The choice of kernel function significantly impacts SVM performance by influencing the model's ability to separate data classes.

• The choice of kernel only affects the training time, not the performance.

List three distance metrics commonly used in KNN.

• Jaccard index

• Hamming distance

• Euclidean distance, Manhattan distance

• Cosine similarity

What is the role of 'k' in the KNN algorithm?

• 'k' indicates the maximum number of features to consider.

• 'k' determines the number of nearest neighbors used for classification or regression.

• 'k' represents the learning rate in the algorithm.

• 'k' is the threshold for data normalization.

Name real-world applications of the KNN algorithm.

• Image compression

• medical diagnosis.

• Weather forecasting

• Stock market analysis

How does KNN handle multi-class classification?

• KNN uses a weighted average of distances for classification.

• KNN only works for binary classification problems.

• KNN requires a predefined threshold for classifying multi-class data.

• KNN classifies by majority voting among the k nearest neighbors for multi-class classification.

In what scenarios would you prefer SVM over KNN?

• When working with small datasets

• When dealing with high-dimensional data, clear class separation, large datasets, or when needing kernel methods.

• In cases with noisy data and outliers

• For problems requiring real-time predictions

What is Machine Learning?

• Writing code to tell the computer what to do

• Teaching computers to learn from data

• Drawing pictures with a computer

• Making games

Which of these is a supervised learning algorithm?

• K-Means

• KNN

• Clustering

• None of the above