<p>about your learning experience to help us improve.</p>

You've completed

Now you have the chance to earn points towards your record of achievement. The specific details of the quiz are listed below. You’ll have one attempt at getting the questions right, so please make sure to check your answers before saving them.
We strongly recommend that you avoid guessing in multiple-answer questions, and only select the answers that you are sure about. Check the [**evaluation of exam performance guidelines**][1] for further details. 

Before you begin, check your browser is compatible. You can find the requirements on our [**FAQs page**][2]. If, for some reason, you get disconnected, try to reconnect within the given time and you’ll still be able to finish the quiz. If you have any other issues, please contact the **openSAP Helpdesk**.

[1]: https://open.sap.com/pages/roa_guidelines#4-evaluation-of-exam-performance
[2]: https://open.sap.com/pages/faq#what-are-the-technical-prerequisites-for-the-weekly-assignments-and-final-exam

ST (MMCP-Storage) - 2024-05-22T16:07:31.154Z)

Assignment Week 4

Getting Started with Data Science

Goal seeking / backward iteration analysis

For what type of analysis do you use the Monte Carlo simulation?

Which of these types of model is an example of double exponential smoothing?

They are used for classification problems only.

They build each model on an independent bootstrap sample (with replacement) from the training data.

They build a forest of regression models.

Which are characteristics of random forest models?

The approach is sensitive to the first split.

They generally examine a single field at a time.

The tree type output can be very hard to interpret.

Which are weaknesses of decision tree algorithms?

What is the temperature forecast for tomorrow if you use a Naïve 1 forecast, where the actual value measured today is 10 Celsius and the actual value measured yesterday is 12 Celsius?

The objective of a linear programming problem is to maximize or minimize some numerical value.

An objective function indicates how much each variable contributes to the value to be optimized in the problem.

The constraints are a set of nonlinear inequalities.

Which statements regarding linear programming are true?

Each model in a boosting ensemble is run independently and outputs are aggregated without preference for any model.

Boosting algorithms create a sequence of models that attempt to correct the mistakes of the models before them in the sequence.

Boosting is more likely to under-fit the model compared to bagging.

In boosting, any instances that previous models misclassified are excluded from the training data for subsequent models.

Which statement regarding boosting is true?

What is the degree of freedom for a contingency table with 4 rows and 2 columns?

Larger values of k can increase the effect of noise.

Larger values of k can make the boundaries between classes less distinct.

How does the choice of the number of nearest neighbors (k) affect the solution?

Neural networks are sometimes referred to as artificial neural networks or multilayer perceptrons.

A neural network consists of two layers - an input layer and an output layer.

A neural network adjusts the weights in order to improve its predictions.

A linear activation function is applied to the sum of the inputs into each node of a neural network.

The back propagation algorithm propagates the error signal back to all the nodes.

Knowledge quiz