Faceted browser

A faceted browser or faceted semantic browser is a user interface which makes use of faceted classification to allow the user to explore by filtering available information. Each facet typically corresponds to the possible values of a property common to a set of digital objects.

Development

The Berkeley Flamenco project pioneered some of this work. While their demos are useful to understand the Flamenco software, links to related projects show the concept in real use, for example to browse NCSU's digital library or the Museo Suomi collections.

Faceted browsing is starting to see some use in rich client applications as well. For an application to be suited to faceted browsing, the digital objects involved must have a common core set of properties. For example, a database of movies will have title, genre, language, length, director, actors and other moderately uniform properties that can be turned into facets to filter on.

Mass market use

Apple iTunes has some limited faceted browsing features. When the browser is activated, the user can choose a genre and simultaneously filter music files by a search term: this combines the genre facet with the search field to narrow the results. Not many facet combinations are possible (for example, a user cannot limit the search results to 5-star rated Rock songs).

Opera's M2 (email client) is a more powerful faceted browser, although again it has arbitrary limitations. A user can easily click to filter their view to see email only from a particular email sender (the email "From" property as the facet) and then choose to filter further by when the email was received (only show emails received last month). The navigation frame offers quick access to facets for who sent the email, what labels it holds, whether it has attachments, and what mailing list it arrived from among others. A drop-down menu offers access to a different set of facets like what time the email was received. A single facet offered in the navigation frame can be combined with a single facet offered in the drop-down menu, but two facets in the navigation frame cannot be combined together. Thus, browsing to "emails sent to the support mailing list from Tim" is not a supported faceted browsing use case.

Facets are often derived by analysis of the text of an item using Entity extraction techniques. This approach permits existing web-pages, product descriptions or articles to have this extra metadata extracted and presented as a navigation facet.

Examples
* Google provides facets at the bottom of their product search to refine by attributes such as price ranges, brands, or vendors.
* Walmart provides faceted search to their shoppers and EBay also allows users to refine their searches with facets.
* [http://www.globrix.co.uk Globrix] crawls existing real estate websites and performs entity extraction to provide facets.

External links

* [http://lucene.apache.org/solr/ Apache Solr]
* [http://www.digital-web.com/articles/user_interface_implementations_of_faceted_browsing/ User Interface Implementations of Faceted Browsing]
* [http://flamenco.berkeley.edu/index.html Flamenco Project]
* [http://mspace.fm mSpace]
* [http://simile.mit.edu/wiki/Exhibit Exhibit - SIMILE]
* [http://www.browseengine.com/ browse engine v1.0]
* [http://drupal.org/project/faceted_search Faceted Search] : A faceted browser for the Drupal content management system (based on Solr).
* [http://slashfacet.semanticweb.org/ Slashfacet] : A semantic web faceted browser that permits the structured browsing of Semantic Web content.
* [http://well-formed-data.net/experiments/elastic_lists/ Elastic lists] : A demonstration of the elastic lists principle for facet browsing.