Compare elastic with Okapi BM25
Read this in "about 1 minute".
Okapi BM25
N — Size of the judged sample
ni — Number of documents in the judged sample containing ti
R — Relevant set size (i.e., number of documents judged relevant)
ri — Number of judged relevant docs containing ti
When there is relevance feedback, it is added to the w.
Okapi BM25F
BM25F considers weighted streams for structured documents.
Compared with Elastic
Improvements
Considering the theory above, relevance feedback and stream weights can be added to the practical elastic scoring function.
The End!
Reference: Stephen Robertson,Hugo Zaragoza.The Probabilistic Relevance Framework: BM25 and Beyond
Author
Typing Theme
Lorem ipsum dolor sit amet, consectetur adipisicing elit. Tempora non aut eos voluptas debitis unde impedit aliquid ipsa.