SE Suite indexing and search standards

The search performed by SE Suite has, by default, a priority hierarchy in each of the indexed field types. However, this does not mean that they will always be returned at the top of the search sorted by priority but that they will have a higher relevance when calculating their search value. This search value is a number calculated by the indexing process that defines the relevance of a record in that search. See below the priority hierarchy defined by the search mechanism:

However, it is also possible to dynamically increase the priority of a token, using a feature known as boost.

To boost a token, use the caret (^) character right after a search token, followed by a numeric value which will be the value of the boost applied to that token. This numeric value will be multiplied by the calculated value of each component record that features it performed by the indexing process.

 
In the examples displayed above, the tokens followed by ^ and a number will receive the boost. In the first example, the softexpert token will receive a boost of 10.

The SE Suite component records will be indexed as to allow searching for tokens in a radicalized way. Thus, it is possible to search for a token regardless of it being singular or plural. However, if desired, it is still possible to choose a form through the previously displayed exact field.

See below examples of tokens and how they are internally indexed. Thus, it is possible to understand how radicalization works.

The search engine ignores certain words which are often used in the language but are ignored when you perform a search. The "null" words are:

"a", "after, "against", "all", "also", "am", "among", "an", "and", "another", "any", "are", "as", "at", "be", "because", "before", "between", "beyond", "both", "but", "by", "during", "either", "every", "even", "from", "for", "he", "her", "hers", "him", "his", "how", "however", "how many", "how much", "if", "in", "is", "less", "like", "many", "more", "mush", "no", "not", "nor", "of", "on", "or", "other", "over", "own", "several", "she", "since", "still", "so", "such", "that", "the", "their", "theirs", "there are", "there is", "these", "they", "this", "those", "through", "to", "too", "under", "what", "when", "where", "whether", "which", "with", "without", "who", "whose", "why", "yet", "your", "yours".

FTS is not sensitive to the use of diacritical signs (acute accent, grave accent, circumflex accent, tilde, cedilla etc.) in the search expressions. For example: when typing resume, resumé or résumé, it will return the same results.

To search and ignore the characters reserved by the search mechanism (?. *, “”, etc.), it is necessary to use a backslash as an escape character (\).

Performing a proper search is of the utmost importance for the desired results to be returned, given simple searches do not always return the expected results. Before assembling the search, it is necessary to know what are the intentions behind it, which will be the base to assemble a search that is appropriate to this particular case.

See below examples of scenarios and queries that may be useful in general.

See below a list of examples of how not to perform a search, that is, assemble a search that does not reflect the interaction of the desired return.