Relevancy tuning options – Google Search Appliance Deployment Governance and Operational Models User Manual
Page 6
6
Data classification/taxonomy
If documents that are to be indexed by the GSA are associated with an enterprise data classification
scheme and/or taxonomy, make sure that ontological information is being indexed by the GSA. Feeding
of ontological information along with the content is not mandatory, but it can enhance the search solution.
The taxonomy can be used in combination with metadata filters to restrict document searches at certain
levels in the overall hierarchy.
Depending on the document type and its source, indexing taxonomy information might require an
integration process that can associate the document classification with the actual document in the index.
A detailed investigation will be needed to see how to associate this information at GSA indexing time.
Entity Recognition
The Entity Recognition feature of the GSA can be used to automatically tag documents with metadata at
index time according to predefined rules contained in text dictionaries or via an XML structure that allows
matches based on terms or regular expressions. Common or standard business terms for your
organization would be great candidates for defining entities. If an enterprise taxonomy has been
identified, this would also be a great candidate for including in Entity Recognition rules. That way, even
though some documents may not be tagged with these terms, the GSA can provide the association at
index time via these rules.
Relevancy tuning options
The Google Search Appliance contains features that empower a search administrator to tweak result
biasing for the entire organization or select search users. Search result ranking can be biased on date,
metadata, source, collections, and feed source.
For
example,
if you have multiple content sources feeding the GSA, with SharePoint being one of them,
and you notice that SharePoint results are undesirably not being ranked high enough in the result set, you
may
choose
to bias the SharePoint content source to improve the ranking of the content.
For
details o
n how to configure a result biasing policy, see the