Inverted indexes, analyzers, relevance scoring with BM25, mappings, shards, query DSL and the ELK stack.
35 cards · basic cards · AI-written, checked twice. Edit anything.
- What is an inverted index?
Show answer
A data structure that maps terms to the documents and positions where they appear, enabling fast full-text search- Why are inverted indexes efficient for text search?
Show answer
They store a mapping from each term to its documents and positions, allowing direct lookup without scanning all documents- What is a postings list?
Show answer
A list of documents and position information for a specific term in an inverted index- What is term vocabulary in inverted indexes?
Show answer
The collection of all unique terms extracted and indexed from documents- What is an analyzer in Elasticsearch?
Show answer
A sequence of character filters, a tokenizer, and token filters that transforms text into searchable tokens- Name the three components of an Elasticsearch analyzer
Show answer
Character filters, tokenizer, and token filters- What does a tokenizer do in an analyzer?
Show answer
Splits text into individual tokens based on rules, producing a stream of tokens for further processing- What is a token filter's role in an analyzer?
Show answer
Modifies, adds, or removes tokens produced by the tokenizer, such as lowercasing, removing stopwords, or stemming- What does a character filter do?
Show answer
Preprocesses input text before tokenization by removing or replacing specific characters or patterns- What does BM25 stand for?
Show answer
Best Matching 25, a probabilistic relevance framework for ranking search results- How does BM25 handle term frequency?
Show answer
It increases scores for more frequent terms but with diminishing returns, so extra occurrences have marginal impact- What is inverse document frequency (IDF)?
Show answer
A measure of how unique a term is across all documents, giving higher weight to rare terms and lower weight to common terms- How does BM25 normalize for document length?
Show answer
It prevents longer documents from gaining unfair relevance advantage by normalizing scores relative to document length- What is the default relevance scoring algorithm in Elasticsearch?
Show answer
BM25, a probabilistic algorithm that ranks documents by relevance to a query- What is a mapping in Elasticsearch?
Show answer
A definition specifying the structure of documents in an index, including field names, types, and analysis settings
FlashKeepers