Quiz on Vector DB

1. What is a vector database primarily used for?

Storing relational tables Storing high-dimensional vectors Storing images as files only Managing text documents only

2. Which of these is a popular vector database?

PostgreSQL Pinecone MongoDB MySQL

3. What is the main advantage of a vector database over a relational database for similarity search?

Supports ACID transactions Efficient high-dimensional similarity searches Stores data in rows and columns Provides relational joins

4. Which metric is commonly used in vector databases for similarity search?

Euclidean distance Hamming distance Cosine similarity All of the above

5. What does ANN stand for in the context of vector search?

Artificial Neural Network Approximate Nearest Neighbor Advanced Node Navigation Automatic Number Normalization

6. Which type of data is commonly stored in vector databases?

Images, text embeddings, audio embeddings Only integers Only strings Only JSON

7. Which query type is vector DB especially optimized for?

Exact match queries Range queries Similarity search / nearest neighbor queries Full-text search only

8. What is an embedding in the context of vector databases?

A SQL table A numeric vector representing data A file format A type of database index

9. Which library is commonly used to generate embeddings for vector databases?

NumPy TensorFlow Matplotlib Pandas

10. Which algorithm is often used for approximate nearest neighbor search?

K-means KD-Tree HNSW (Hierarchical Navigable Small World) BFS

11. What is cosine similarity used for?

Measuring distance in kilometers Measuring angle similarity between vectors Calculating database latency Sorting numeric tables

12. Which of these is a cloud-native vector database?

Milvus Pinecone MySQL Redis

13. What is a shard in vector databases?

A single vector A partition of data for scaling A query type A metric type

14. What kind of indexing improves vector search speed?

Hash-based indexing Tree-based indexing Full-text indexing No indexing

15. Which of the following is true about vector embeddings?

They are always 2D They capture semantic meaning They are not numeric They cannot be stored in databases