1. Which measure is least affected by outliers?

Mean Median Range Variance

2. Variance measures:

Central tendency Data spread Probability Data size

3. A normal distribution is:

Skewed Uniform Symmetric Discrete

4. Expected value represents:

Most likely outcome Long-run average Maximum value Median value

5. Which is a continuous variable?

Number of students Test score Height Category

6. Best chart to show relationship between two numeric variables?

Histogram Boxplot Scatterplot Pie chart

7. Boxplots are useful for identifying:

Correlation Outliers Trends Time series

8. Logistic regression is used when output is:

Continuous Binary Clustered Random

9. Data cleaning improves:

File size Model accuracy Data duplication Noise

10. K-means is an example of:

Classification Regression Clustering Forecasting

11. Which SQL clause filters rows?

SELECT WHERE JOIN ORDER BY

12. Pandas is primarily used for:

Visualization Data manipulation Web apps Networking

13. NumPy specializes in:

Text processing Numerical arrays Databases APIs

14. Relational databases store data in:

Documents Tables Graphs Images

15. Generative AI is designed to:

Classify data Create new content Store data Encrypt files

16. A major limitation of generative AI is:

Speed Hallucinations Storage Connectivity

17. NLP focuses on:

Images Language Robotics Sensors

18. LLMs are trained mainly on:

Sensor data Large text datasets Images only Hand-coded rules

19. Which dataset is used for final evaluation?

Training Validation Test Labeled

20. Unsupervised learning uses:

Labeled data No labels Rewards Rules

21. Deep learning relies on:

Linear equations Neural networks Decision tables SQL queries

22. Predicate logic is:

Probability-based Rule-based Neural-based Random

23. Bayesian networks represent:

Linear systems Probabilistic relationships Databases Text corpora

24. Algorithmic bias often comes from:

Hardware Biased training data Encryption Storage limits

25. Hallucinations refer to:

Security breaches Confident false outputs Slow models Data loss

26. Structured data is:

Free text Images Tabular Audio

27. Binary numbers use base:

2 8 10 16

28. Data wrangling involves:

Data collection only Cleaning and transforming Visualization Deployment

📊 Data Science & AI – Full Practice Quiz (All Topics)

Probability & Statistics Foundations

Data Analysis & Statistics for AI

Tools for Data & AI

AI Basics

Machine Learning

Perception, Representation & Reasoning

Privacy, Ethics & Data Literacy

🧠 Flashcards – Key Concepts