Fuzzy Search

A fuzzy query returns results that contains words that are similar to your query string. You can use fuzzy queries when you are not sure how to spell some of the words that you want to query for.

You can use the following formats for a fuzzy query:

  • Text=<my query text>DREFUZZY(FuzzyQueryText)

    For example:

    http://localhost:9100/action=Query&Text=best selling author DREFUZZY(Rowlling)
  • Text=DREFUZZY(FuzzyQueryText)

    For example:

    http://localhost:9100/action=Query&Text=DREFUZZY(Caroll Jabberwalky)

You can also use Boolean Operators within fuzzy queries.

  • Text=DREFUZZY(FuzzyQueryText OPERATOR OtherFuzzyQueryText)

    For example:

    http://localhost:9100/action=Query&Text=DREFUZZY(Caroll AND Jabberwalky)

Adjust the Tolerance Level of Fuzzy Queries

By default, DREFUZZY internally determines an appropriate tolerance level when it judges if words in a result document are similar enough to the words that are specified in the query. However, in exceptional circumstances you can postfix DREFUZZY with a numerical value to adjust this tolerance level:

DREFUZZYN(FuzzyQueryText)

This value operates like a slider that increases or decreases the tolerance level. If this value is high, IDOL Content Component returns more words as potential matches for the query terms. If the value is low, IDOL Content Component returns fewer words as potential matches.

NOTE: Micro Focus does not recommend specifying a value higher than six because it might result in fuzzy matching that is so flexible that results might be unrelated to the query.

Example:

action=Query&Text=DREFUZZY1(Caroll Jabberwalky)
action=Query&Text=DREFUZZY3(Caroll Jabberwalky)

The second query in this example is more tolerant in accepting matches for the specified words than the first query, which means that it might return more results.