Relevance ranking is a search process which uses fuzzy logic to retrieve as many records as possible. A search on "black" and "white" might retrieve 20 records which include both terms. There may also be a further six records which contain only "black" and 15 which contain "white". The records that contain both words are the most relevant to the search and will be listed first. If there are no records which exactly match an enquiry, a relevance ranked search will still retrieve a list which matches part of the search text.
a programmed algorithm applied to online search engines which results in displaying first the retrieved records that are most likely to be relevant to the search query and decreasing in order of likely relevance. Among the more commonly used ranking criteria are: term frequency (the number of times the term occurs in the document), document frequency (the number of documents containing the term), proximity of query words, variant word forms (stemming), and document length.