User Guide
172 Getting Results with Novell Web Services
Getting Results with Novell Web Services
103-000133-001
August 29, 2001
Novell Confidential
Manual 99a38 July 17, 2001
To illustrate how these criteria work, consider the following examples:
Words in bold face are more relevant than regular words.
Words contained in the <Title> tag are more relevant than words
contained within the <body> tag.
Words contained in the Keywords and Description META tags are more
relevant than content words.
Words contained within the <A HREF=> tag used for creating links are
less relevant than words outside of this tag.
A document containing a specified search term multiple times is more
relevant than a document that contains the search term only once.
A word within a 36-point body text is more relevant than within 4-point
footer text.
Documents returned from a query that is weighted at 100% is more
relevant than a 50% weighted query. This is normally used in multiquery
searches where each query has a specified weight, as in query=0;
weight0=; query1=; weight1=.
Weighted Queries
A weighted query is used anytime you want to modify the order or relevance
of certain hits in a user’s normal search results list or when you want to add
additional search results users might not have identified in their queries.
The following query parameters can be combined to identify a single search
query item:
&filter#=
&filteroperator#=
&operator#=
&query#=
&weight#=
Multiple query items can also be sent as part of a single search request.
You can use this feature to provide profile-enhanced search requests. For
example, the following query returns French product downloads higher up in
the search results list but does not eliminate results of any other language
downloads:
&query0=product+downloads&weight0=100&query1=/
product=french&weight1=90