Exam DP-100 topic 1 question 31 discussion

should be linear regression

Correct answer is C : It builds each regression tree in a step-wise fashion, using a predefined loss function to measure the error in each step and correct for it in the next. Thus the prediction model is actually an ensemble of weaker prediction models. reference:https://docs.microsoft.com/en-us/azure/machine-learning/algorithm-module-reference/boosted-decision-tree-regression

The correct answer is Boosted Decision Tree Regression. Boosted Decision Tree Regression is an algorithm that reduces the variance between actual and predicted values by iteratively combining multiple weak learners (decision trees) to create a stronger, more accurate model. This helps minimize the differences between the predicted values and the actual values by focusing on the errors made by previous models. Linear regression, while useful for linear relationships, does not specifically focus on reducing variance in the way ensemble methods like boosting do.

D is linear and reduces the mean squared error which is the variance between actual and predicted. Regression trees do that too but are not linear

Linear regression computes the linear relationship between a dependent variable and one or more independent features. It uses the ordinary least squares method to minimize the sum of the squared errors between the actual and predicted values.

D. Linear Regression

Question keyword 'develop a linear regression model'. Answer keyword 'Linear regression' make (D) is correct answer. https://learn.microsoft.com/en-us/azure/machine-learning/component-reference/linear-regression?view=azureml-api-2 Quote: 'create a linear regression model' , 'Linear regression is a common statistical method, ...'

I think the question is poorly asked, since the algorithme used depends mainy on the data. The answer is linear regression, since Poisson regression is actually a form of generalized linear modeling and is often considered a non-linear regression model due to its underlying mathematical structure. Also Boosted decision tree, in a non linear regression as well.

Boosted decision tree regression a non-linear regression model.

Boosting will definitely reduce the variances

Answer should be C. Reference: https://www.kdd.org/kdd2016/papers/files/adf0653-poyarkovA.pdf

Decision trees are non linear. Unlike Linear regression there is no equation to express relationship between independent and dependent variables

guys I know some of these questions are crap but read carefully it says "assess various algorithms.", which means a boosting algorithm like Boosted Decision Tree Regression makes sense.

ChatGPT answer: Option D: Linear Regression, is an algorithm that can help to reduce the variances between actual and predicted values in a linear regression model.

Boosted Decision Tree Regression is an algorithm that reduces the variances between actual and predicted values. Linear regression aims to find the best linear relationship between the independent and dependent variables, while Fast Forest Quantile Regression is a regression algorithm that can provide estimates of conditional quantiles of the response variable. Poisson Regression is used when the dependent variable is a count or frequency, and the relationship between the dependent and independent variables is modelled using a Poisson distribution.

D. Linear Regression is an algorithm that reduces the variances between actual and predicted values. Linear regression is a type of regression analysis that finds the linear relationship between a dependent variable and one or more independent variables. The goal of linear regression is to minimize the sum of the squared residuals (the differences between the predicted and actual values), which is a measure of the variance between the predicted and actual values. Therefore, linear regression aims to reduce the variances between actual and predicted values.

Boosted Decision Tree Regression is an algorithm that reduces the variance between actual and predicted values. It combines the outputs of multiple decision trees, which helps to generate a more accurate prediction by reducing the variance in the model. Linear Regression, on the other hand, assumes a linear relationship between the input features and the target variable, but it does not specifically reduce the variance between actual and predicted values. Once again, I apologize for any confusion caused by my initial response and hope that this clarification is helpful.