Question 1

A healthcare AI model trained on adult patient data performs poorly when deployed in pediatric clinics. What fundamental issue does this scenario reveal?

Options :

A : Data normalization failure

B : Overfitting to noisy labels

C : Poor generalization across demographic domains

D : Redundant feature extraction

Answer: C

Question 2

A financial model uses income and credit score data for predicting loan approval. The income data has values ranging from 20K to 500K, while the credit score data ranges from 300 to 850. What data preprocessing technique is required to handle these features?

Options :

A : Principal Component Analysis (PCA)

B : Imputation for missing data

C : One-hot encoding for both features

D : Feature scaling (standardization or normalization)

Answer: D

Question 3

A financial services company is developing an AI-powered credit scoring system. The model needs to classify customers into low-risk, medium-risk, and high-risk categories based on historical financial data. Which type of AI learning should be used?

Options :

A : Supervised Learning, as it utilizes labeled data to train the model on past credit scores and customer profiles

B : Federated Learning, since it allows multiple institutions to train the model without sharing sensitive data

C : Unsupervised Learning, as it does not require historical labels for training

D : Deep Learning, as it requires large-scale neural networks to process customer financial history

Answer: A

Question 4

A city-wide traffic optimization AI performs inconsistently depending on time of day and regional sensor coverage. Engineers note the model struggles in low-data or fluctuating conditions. What approach would most effectively address this reliability issue?

Options :

A : Static routing fallback

B : One-hot encoding of sensor IDs

C : Ensemble modeling with diverse temporal and spatial inputs

D : Quantization-aware training

Answer: C

Question 5

A financial institution wants to automate the approval of mortgage applications based on applicant income, credit score, and loan amount. The system needs to classify each application as "Approved," "Pending Review," or "Rejected." Which AI model should be used?

Options :

A : Clustering Model

B : Classification Model

C : Generative Model

D : Regression Model

Answer: B

Confidently Practice Online with Free AI-Essentials Exam Cram

Buying Options: