Practise vocabulary for semantic and vector search: dense retrieval, embeddings, FAISS, ANN, hybrid search, and vector databases.
0 / 10 completed
1 / 10
___ search uses vector representations of text to find semantically similar documents, even when they don't share exact keywords.
Semantic search uses dense vector embeddings (from models like BERT, sentence-transformers) to represent meaning. Two sentences with different words but similar meaning have similar vectors, enabling retrieval based on conceptual similarity.
2 / 10
In vector search, a text is converted to a fixed-length numeric vector called an ___.
An embedding is a dense vector representation of text (or images, audio) in a high-dimensional space, produced by a neural model. Semantically similar content has similar vectors — measured by cosine similarity or dot product.
3 / 10
FAISS (Facebook AI Similarity Search) is a library for efficient ___ search — finding the closest vectors to a query vector at scale.
FAISS provides ANN (Approximate Nearest Neighbour) search, finding the most similar vectors without exhaustively computing distances to all stored vectors. ANN trades a small accuracy loss for dramatic speed improvements at billion-vector scale.
4 / 10
___ search combines sparse (BM25 keyword) and dense (vector) retrieval to leverage both lexical and semantic matching.
Hybrid search merges BM25 keyword results and vector similarity results using a fusion algorithm (RRF — Reciprocal Rank Fusion). This outperforms either approach alone by capturing both exact keyword matches and semantic similarity.
5 / 10
A ___ database (e.g., Pinecone, Weaviate, Qdrant, pgvector) is purpose-built for storing and querying high-dimensional embedding vectors.
Vector databases store embeddings and support efficient ANN search, filtering by metadata, and hybrid retrieval. They're the foundation of RAG (Retrieval-Augmented Generation) systems and recommendation engines using semantic similarity.
6 / 10
Liam, during a code review of the new user onboarding flow, says: 'I'm using semantic search to find relevant documentation based on the user's problem description. It's much better than just searching for keywords like 'password reset'.' Which of the following best describes Liam's approach?
Liam correctly describes semantic search. Traditional keyword-based search relies on exact matches, while semantic search leverages vector representations to understand the *meaning* of the query and find semantically similar documents – even if they don't use identical keywords. The other options misinterpret or oversimplify the core principle of semantic search.
7 / 10
Sarah in the Slack channel #dev-ops is discussing a recent issue with slow query times. A senior engineer responds: 'We've been experimenting with integrating dense vector embeddings into our search index to improve recall. We're using FAISS for efficient similarity searches.' What is Sarah and the engineer primarily referring to?
Sarah and the engineer are discussing vector embeddings. Dense vector representations are used to represent textual data in a way that allows semantic search algorithms to understand the *meaning* of the text – this is central to their discussion about improving recall. FAISS is a library specifically designed for efficient similarity searches using these vectors.
8 / 10
Mark needs to describe the architecture of his new search feature in a PR description. He writes: 'We're utilizing a hybrid retrieval system combining BM25 for initial keyword filtering with vector embeddings for refined semantic matching. This approach aims to balance precision and recall.' Which component is Mark MOST likely referring to?
Mark's description highlights a hybrid retrieval system. Combining BM25 (sparse) and vector embeddings (dense) is a common strategy in semantic search. BM25 performs initial keyword filtering to reduce the search space, while vector embeddings provide more nuanced semantic matching for improved recall – this combination maximizes both precision and recall.
9 / 10
David is preparing a standup update. He says: 'We've migrated our documentation search to a Pinecone vector database. This allows us to quickly find related articles based on the *intent* of the user's query, not just exact keyword matches.' What is the primary benefit David highlights?
David is emphasizing the core benefit of using a Pinecone database for semantic search. Pinecone's purpose-built architecture allows it to efficiently store and query high-dimensional embedding vectors – which are central to understanding the *intent* behind user queries and performing similarity searches based on meaning.
10 / 10
Emily is troubleshooting performance issues with a semantic search application. She notices that query latency is high. Which of the following factors is MOST likely contributing to this problem?
The size of embedding vectors plays a crucial role. Smaller vectors may lack the dimensionality needed to fully represent the semantic nuances of the text, leading to less accurate similarity calculations and increased latency. Optimizing the index for query patterns is also critical; however, this is a secondary concern compared to vector size.
What will I learn from the "Semantic Search Vocabulary" exercise?
Practise vocabulary for semantic and vector search: dense retrieval, embeddings, FAISS, ANN, hybrid search, and vector databases.
Is this exercise free to use?
Yes. Every exercise on CoderSlingo, including this one, is free to use with no account, sign-up, or paywall required.
How many questions are in this exercise?
This set contains 10 multiple-choice questions, each with a detailed explanation shown after you answer.
Do I need to create an account to track my progress?
No account is required. Your progress bar and score reset each time you reload the page, but you can retry the exercise as many times as you like.
Who is this Search Engineering Language exercise for?
This exercise is built for IT professionals and non-native English speakers who need to read, write, and discuss search engineering language topics confidently at work.
What happens if I answer a question incorrectly?
You will see the correct answer highlighted along with a detailed explanation of why it is correct -- so every wrong answer becomes a learning moment, not just a lost point.
Can I retry this exercise?
Yes -- click "Try again" on the results screen at any time to reset your score and go through all the questions again.
How long does this exercise take to complete?
Most learners finish all 10 questions in under 10 minutes, since each question is answered by clicking a single option.
Where can I find more Search Engineering Language exercises?
See the full Search Engineering Language exercises hub for more vocabulary drills on this topic.
Is this exercise mobile-friendly?
Yes -- the exercise works on any device with a modern browser, including phones and tablets, with no app download required.