Exploring Linear Classification Techniques

What is the primary goal of linear classification?

• To create a non-linear decision boundary for complex datasets.

• To classify data points based on their distance from the nearest neighbor.

• To maximize the distance between data points and the origin.

• To find a linear decision boundary that separates different classes.

Define univariate linear regression.

• Univariate linear regression is a method to model the relationship between one independent variable and one dependent variable using a linear equation.

• Univariate linear regression predicts categorical outcomes using a linear model.

• Univariate linear regression is a method to analyze time series data.

• Univariate linear regression involves multiple independent variables.

What is the difference between univariate and bivariate regression?

• Univariate regression analyzes multiple dependent variables; bivariate regression analyzes one.

• Univariate regression uses two independent variables; bivariate regression uses one.

• Univariate regression is used for categorical data; bivariate regression is used for continuous data.

• Univariate regression uses one independent variable; bivariate regression uses two.

Explain the concept of multivariate linear regression.

• It only applies to categorical data without any numerical variables.

• It predicts multiple dependent variables using a single independent variable.

• It analyzes the relationship between two dependent variables without any independent variables.

• Multivariate linear regression models the relationship between one dependent variable and multiple independent variables.

What is regularized regression and why is it used?

• Regularized regression eliminates the need for data preprocessing.

• Regularized regression is used to prevent overfitting by adding a penalty to the regression model.

• Regularized regression is used to increase model complexity.

• Regularized regression focuses solely on maximizing R-squared.

How does logistic regression differ from linear regression?

• Logistic regression predicts probabilities for classification, while linear regression predicts continuous values.

• Logistic regression is used for time series forecasting.

• Linear regression can only handle binary outcomes.

• Logistic regression requires normally distributed data.

What are the assumptions of linear regression?

• Normality, independence, and non-linearity

• Linearity, independence, and non-normality

• The assumptions of linear regression are linearity, independence, homoscedasticity, normality, and no multicollinearity.

• Heteroscedasticity, multicollinearity, and independence

Describe the role of the cost function in linear regression.

• The cost function only applies to non-linear regression models.

• The cost function is used to increase the complexity of the model.

• The cost function measures the error between predicted and actual values, guiding the optimization of model parameters.

• The cost function predicts future values directly without measuring error.

What is the purpose of the gradient descent algorithm?

• To sort data in ascending order.

• To minimize a function and find optimal parameters.

• To maximize a function and find the worst parameters.

• To calculate the average of a dataset.

How can you assess the performance of a linear regression model?

• Ignore residuals and only check predictions.

• Use R-squared, MAE, MSE, RMSE, and residual plots.

• Focus solely on the training accuracy.

• Use only the mean and median values.