C_PAII10_35 Online Practice Questions and Answers

For data and performance management purposes, the dataset to be analyzed must contain a variable that serves as a key variable. Two cases should be considered:

Note: There are 2 correct answers to this question.

A. If the initial dataset does not contain a key variable, a variable index Kx Index is automatically generated by the application. This will correspond to the row number of the processed data.

B. If the file contains one or more key variables, they are not recognized automatically. You must specify them manually in the data description.

C. the user who runs SAP Predictive Analytics is also the one used to install the software

Once the modeling parameters are defined, you can generate the model. Then you must validate its performance using the predictive power.

Note: There are 2 correct answers to this question.

A. If the model is sufficiently powerful, you can analyze the responses that it provides in relation to your business issue apply it to new datasets

B. Otherwise, you can modify the modeling parameters in such a way that they are better suited to your dataset and your business issue, and then generate new, more powerful models.

C. The Progress Bar displays the progression for each step of the process. It is the screen displayed by default.

D. The Detailed Log displays the details of each step of the process.

Why Describe the Data Selected? In order for Automated Analytics features to interpret and analyze your

data, the data must be described.

To put it another way, the description file must specify the nature of each variable, determining their:

Note: There are 2 correct answers to this question.

A. Storage format: number

B. Type: continuous, nominal, ordinal or textual

C. A Java Runtime Environment installed on the client machine to connect to the server

D. A valid SAP Predictive Analytics License

You built a classification model using SAP HANA data with Automated Analytics. How will you make the model available to Model Manager?

Note: There are 1 correct answers to this question.

A. Generate Source Code

B. Apply Model

C. Save Model

D. Load a Model

Predictive Power, Prediction Confidence and Model Graphs On the model graph plot: Note: There are 2 correct answers to this question.

A. Produce deep analysis of the data using different visualization techniques, such as scatter matrix charts, parallel coordinates, cluster charts, and decision trees.

B. Of the validation dataset of the perfect model and that of the random model". As the curve of the generated model approaches the curve of the perfect model, the value of the predictive power approaches 1.

C. Of the training and validation datasets the training dataset and that of the validation dataset "divided by "the area found between the curve of the perfect model and that of the random model"

The term model carries many different meanings depending on its field of application. In data mining, a model describes and explains the relationships that exist between input data (one or more target variablest allows one to predict and explain phenomena, or to describe them. To quote George E.P. Box, "All Models Are Wrong But Some Are Useful" from "Robustness is the Strategy of Scientific Model Building" in Robustness in Statistics. Eds., R.L. Launer and G.N. Wilkinson, 1979, Academic Press. A model that has satisfactory performance is one that possesses both:

Note: There are 2 correct answers to this question.

A. High explanatory power, that is, sufficient capacity to explain the target variable. This explanatory power is indicated by the predictive power of the model.

B. High robustness, that is, sufficient capacity to repeat the same performance on new datasets containing observations of a similar nature to the training dataset. This explanatory power is indicated by the prediction confidence .

C. Both of these ports must be accessible from the client machine. For example, network firewalls should allow communication on these ports between client machine and the servers.

To Validate the Model Generated :

Note: There are 2 correct answers to this question.

A. Verify the Predictive Power Automated Analytics User Guides and Scenarios Modeler If the performance of the model meets your requirements, go to Step 3 - Analyzing and Understanding the Model Generated. Otherwise, go to the procedure To Generate a New Model.

B. You can also check other indicators provided in addition to KI and KR during the model generation. For example, you could view the total elapsed time required to generate the model and information on the standard error rate.

C. A web server such as Apache Web Server or Windows Internet Information Services (IIS).

To Define the Risk Mode Parameters:

Note: There are 3 correct answers to this question.

A. In the field Risk Score, enter the score you want to associate with a good/bad odds ratio.

B. In the field for good/bad odds ratio of, enter the ratio.

C. Indicate the increase of score points needed to double the odds in the field Points to double odds.

D. Click the button View Score Table to display the table of scores associated with the corresponding good/bad odds ratio.

E. The existing R is uninstalled and the registry entries and the R installation folder are removed from the machine

There are three ways to set the threshold using the displayed slide bar: Note: There are 3 correct answers to this question.

A. By selecting the percentage of population to target if the population is sorted by descending order of score

B. By selecting the percentage of positive observations you want to detect

C. By selecting the score used to differentiate positive observations from negative ones considered positive, on the contrary any observation with a score below the threshold is considered negative.

D. It is recommended to install the application as a Windows service on a production environment.

Understanding the Decision Tree Panel The panel Decision Tree is split into three parts: Note: There are 3 correct answers to this question.

A. The decision tree itself, which is displayed in the upper section of the panel

B. Two tabs located in the left bottom part of the panel provide you with information on the nodes and with the profit curve corresponding to the current decision tree

C. A navigator allowing you to visualize what part of the tree you are studying is displayed in the right bottom part of the panel.\

D. A consequence of 1 is that this metric is resistant to outliers: when there are a few occurrences of the target with very high values with respect to the rest of the target value distributions, the notion of normal profit is not impacted.