MIS447 Quiz2-Exploring Machine Learning Concepts

What is one-hot encoding and why is it used?

• One-hot encoding is a method for generating random numbers.

• One-hot encoding is used to reduce the size of numerical data.

• One-hot encoding is a technique for image compression.

• One-hot encoding is a method to convert categorical variables into a binary format for machine learning.

What is the advantage of k-fold cross validation?

• It guarantees a perfect model performance.

• It provides a more reliable estimate of model performance by using multiple training and validation sets.

• It eliminates the need for any validation data.

• It reduces the computational cost of training the model.

Why is stratified sampling often preferred over purely random sampling when creating a test set in machine learning?

• It ensures that the test set is smaller than the training set.

• It guarantees that the test set contains an equal number of instances from each class.

• It ensures that the test set is representative of the overall population by maintaining the proportions of specific subgroups (strata).

• It reduces the computational cost of training the model.

When reshaping a NumPy array using the reshape() function, what will happen if you try to reshape an array into an incompatible shape?

• The array will be reshaped successfully.

• An error will occur indicating an incompatible shape.

• The original array will be modified to the new shape.

• The area will automatically adjust to fit the new shape.

What does following code snippet do?

• Plots a line graph of total bills over different days.

• Creates a box plot showing the distribution of total bills across different days.

• Generates a histogram of total bills.

• Displays a box plot of tips versus total bills.

What are the main differences between supervised and unsupervised learning?

• Unsupervised learning uses a smaller dataset than supervised learning.

• Supervised learning is faster than unsupervised learning.

• Supervised learning can only be used for classification tasks.

• Supervised learning requires labeled data, while unsupervised learning does not.

Explain the concept of support vector machines.

• Support Vector Machines are unsupervised learning models that cluster data points without labels.

• Support Vector Machines are supervised learning models that find the optimal hyperplane to separate different classes in high-dimensional space.

• Support Vector Machines are primarily used for regression tasks in low-dimensional space.

• Support Vector Machines rely on decision trees to classify data points.

What is the role of the K-nearest neighbour algorithm?

• The K-nearest neighbour algorithm classifies or predicts values based on the closest training examples in the feature space.

• The K-nearest neighbour algorithm is used for optimizing database queries.

• The K-nearest neighbour algorithm generates random predictions without using training data.

• The K-nearest neighbour algorithm only works with linear data distributions.

Describe the steps involved in a typical machine learning workflow.

• Define the problem, collect data, preprocess data, split data, select model, train model, evaluate model, tune hyperparameters, deploy model, monitor model.

• Collect data only

• Evaluate model before selecting model

• Deploy model before training

What is GridSearchCV used for in model tuning?

• GridSearchCV is used for feature selection.

• GridSearchCV is used for data preprocessing.

• GridSearchCV is used for hyperparameter tuning in model selection.

• GridSearchCV is used for model evaluation.