About this document
Vector Space Model and BM25 Scoring Explained by Mariem El Mechry is a document available to read on EtoBox.
The document discusses how documents are ranked in Elasticsearch. It explains that Elasticsearch first filters documents that match the query boolean, then calculates a score for remaining documents to determine the ranking. The default scoring algorithm is BM25, which considers term frequency, inverse document frequency, and field length to calculate relevance scores. BM25 improves upon simple term frequency by introducing a saturation parameter to diminish scores for frequent terms.
- Author
- Mariem El Mechry
- Language
- EN