Use this page to modify the Search terms that will be used to find content in this particular search group and site.

InterPublish employs up to three kinds of search terms to ensure you locate only the documents you are interested in - no more and no less.

1) Site search terms

In the initial site search, Interpublish will use your terms to identify an initial set of documents that match your criteria.  It will  refer to the site setup information (under Manage Sites) to help it extract key metadata from the HTML (Title, Author, Date, URL, Text) and write these into the InterPublish database together with a PDF version of the page or document. 

The Site search terms are the words or phrases you tell InterPublish tol supply to the target site's search engine to produce an initial candidate list of relevant documents.  At least one site search term must be present.

The site search terms you enter in a list are logically related to each other by an implied logical OR operator. In the following example the site will be searched for documents containing 'age discrimination' OR 'aged care' OR 'ageing' OR 'alzheimer' OR 'carers'.  If you wish you can state this relationship explicitly by including 'or:' before a search term but since 'or' is the default behaviour you can also leave it off. Unlike the case with filters (below) the AND operator is not available for site searches.

More advanced website search engines can be directed to limit the scope of their search to certain metadata fields.  By default, a site search will scan all 'keywords' but (depending on the search engine) it may be possible to direct it to search only particular fields such as 'author' or 'tag'.  If this would be helpful, you can prefix the search term with the following qualifiers:

• author:
• tag:

If you have specified a qualifier but a search engine doesn't support it, your qualifier will be ignored and a keyword search performed for the term.

The following example shows a list of site search terms, some of which are modified by qualifiers.

example 1

2) Positive filters

Filters are a second stage refining procedure applied by Interpublish to the document metadata it captured using site search terms (above). Filtering may be thought of as shortlisting the documents.  To make the shortlist a candidate document must match the Positive filter terms.  Positive filters are not mandatory, but they are highly recommended - they narrow the document set down to those that better match your criteria.

Like Site search terms, filter words and phrases are logically related to each other by an implied Boolean OR operator, however, a filter can also take an AND operator as shown in the following example:

example 2

Because a fillter is working with more structured metadata extracted during the initial search and stored in the InterPublish database, there are more options available for filtering based on metadata fields, specifically: author, title and text (the body text of the document).  The following qualifiers can therefore be used when specifying a filter:

• author: 
• title: 
• text: 

The following example shows how logical operators and field qualifiers can be combined to shortlist documents that were selected by the initial site search.

example 3

3) Negative filters

Negative filters are also post-search filter terms applied by Interpublish on the documents selected by the site search.  For a document to survive the negative filter it must NOT match the negative filter terms. Negative filters are optional, but highly recommended.

Negative filters behave like Positive filters in terms of their logical operators and field qualifiers.  The difference is that whereas positive filters describe terms that must be present, negative filters describe terms that must not be present.