BSCS 4-4 - Elective 4 (Machine Learning) - Quiz #2 - 5-8-2024

What is the primary goal of regression in machine learning?

• To classify data into discrete categories

• To predict a continuous target variable

• To cluster data into groups

• To reduce the dimensionality of the data

In the context of linear regression, what does the term "ordinary least squares" refer to?

• A method for classifying data points

• A technique for reducing the complexity of a model

• 
  A strategy for increasing the margin between classes

• The most common method to train the linear regression equation

A researcher is studying the relationship between the number of hours studied and exam scores. What type of regression should they use?

• Multiple linear regression

• Logistic regression

• Simple linear regression

• Polynomial regression

Which assumption must not be violated in linear regression analysis concerning error terms?

• Homoscedasticity

• Independence of observations

• Non-normality

• Multicollinearity

What are some significant constraints or drawbacks associated with using simple linear regression for predictive modeling and statistical analysis?

• It can only model non-linear relationships

• It assumes that the relationship between variables is always indirect

• It cannot determine causation, only correlation

• It uses multiple coefficients for prediction

What are the recommended practices for preparing and transforming data to optimize its suitability for analysis using linear regression techniques?

• Data should be transformed to remove non-linearity

• Data should be clustered before applying regression

• Input and output variables should have non-Gaussian distributions

• Outliers should be added to increase model robustness

What is the key difference between simple linear regression and multiple linear regression?

• Simple linear regression is more accurate than multiple linear regression

• Simple linear regression is more computationally efficient than multiple linear regression

• Multiple linear regression can only be used for classification tasks, while simple linear regression is used for regression tasks

• Simple linear regression has one input variable, while multiple linear regression has multiple input variables

During a project, a neural network engineer is tasked with improving a model that classifies images into different categories. The model is currently experiencing high variance. Which approach should the engineer consider to address this issue?

• Increasing the complexity of the model by adding more layers

• Simplifying the model by reducing the number of neurons

• Implementing dropout techniques during training

• Decreasing the batch size during training

Which method is primarily used for finding the best-fit line in linear regression?

• Gradient descent

• Ordinary Least Squares

• Maximum likelihood estimation

• Ridge regression

How does the Naive Bayes algorithm calculate the posterior probability P(c | x)?

• P(c|x) = P(c) * P(x|c) / P(x)

• P(c|x) = P(x) * P(c|x) / P(c)

• P(c|x) = P(x) * P(c) / P(x|c)

• P(c|x) = P(c) * P(x) / P(x|c)