Question 1

Which of the following scenarios represents a binary classification problem?

Options :

A : Predicting the probability of different stock prices

B : Forecasting next month’s sales in dollars

C : Determining if a transaction is fraudulent or legitimate

D : Grouping customers into three income tiers

Answer: C

Question 2

A tech company is developing an AI tool for educational institutions in Europe, which includes emotion recognition technology to analyze student engagement. Under the EU AI Act, what should the company consider?

Options :

A : Only deploy the technology after collecting sufficient data on student reactions.

B : Ensure that the tool is easily accessible to all students and teachers.

C : Market the technology broadly as a tool for all learning environments.

D : Avoid deploying the tool as emotion recognition in educational settings is likely prohibited under the AI Act.

Answer: D

Question 3

A financial institution is developing a credit risk model using a machine learning algorithm. The data science team decides to split their dataset into three parts: training, validation, and test sets. What is the primary purpose of the validation set in this setup?

Options :

A : To finalize the model by evaluating it on unseen data

B : To select the best performing model among multiple algorithms

C : To adjust model parameters to fit the training data

D : To test the model’s performance under different scenarios

Answer: B

Question 4

An insurance firm uses a predictive model to detect fraud. In the interest of balancing sensitivity and specificity, which of the following challenges might the firm face?

Options :

A : Maximizing sensitivity without increasing the false-positive rate

B : Reducing the base rate to ensure fair outcomes

C : Ensuring predictive rate parity and equal opportunity simultaneously

D : Applying demographic parity to improve prediction quality

Answer: A

Question 5

A justice system uses an AI model to predict failure-to-appear (FTA) rates in court. To ensure fairness without explicitly using protected group data, what could be a potential unintended outcome?

Options :

A : The model underestimates FTA for all demographics

B : The model’s accuracy decreases for high-risk groups

C : The model’s reliance on correlated variables increases potential bias in predictions

D : The model achieves demographic parity but lacks predictive accuracy

Answer: C