Fuzzy search (general inaccuracy)

Modified on Tue, 14 Oct at 11:35 AM

By placing one or more tilde symbols ~ in front of a keyword, a fuzzy search is performed. Fuzzy means that not all letters of the keyword have to match exactly with the words contained in the articles, but rather a certain degree of tolerance (fuzziness) is allowed when comparing characters.

Examples:

Tilde	Meaning
~CLOUD	Maximum Levenshtein distance = 1 Finds among others: CLOUD CLAUDE LOUD
~~OTO	Maximum Levenshtein distance = 2 Finds among others: PHOTO OUTO OTTO
~~WINE	Maximum Levenshtein distance = 2 Finds among others: WINE WHINE WIND
~~GADDAFI	Maximum Levenshtein distance = 2 Finds among others: GADAFFI GADAFI GADDAFI GHADAFFI GHADAFI

Stem search (stemming or lemmatisation)

In a stem search, the keyword is reduced to its basic form and then terms are searched for derived from this root word.

The STEM function is formulated as follows in a search:

STEM.k(PHRASE)

The parameter k is used to specify the type of stemming to be applied to the specified phrase during the search.

Function	Meaning
STEM.0(Phrase) or NOSTEM(Phrase)	No stemming should be performed on the specified phrase. The phrase is searched for as specified.
STEM.1(Phrase) or STEM(Phrase)	The search for phrases should be performed using normal stemming (e.g. a search for “tree” will also find articles containing “trees”).
STEM.2(Phrase) or SYNO(Phrase)	The search for phrases should be performed using lemmatisation (e.g. a search for “go” will also find articles containing “went”).

Note:

STEM.0 is the default setting for all keywords in a syntax search and therefore does not need to be specified explicitly.

STEM.1 is the default setting in simple and extended searches.