Have you tried Dieselpoint Search™ yet?


Faceted Navigation

Dieselpoint Search provides the full array of full-text search capabilities expected of any high-end search engine — keyword search, linguistic capabilities, and fast response time.

It also provides a radically new way to search and navigate through data and documents. Dieselpoint’s Search and Navigation™ is a way of using the attributes of a data set to build menus that allow a user to browse in an intelligent way. Sometimes called faceted navigation or guided search, developers can use this data-about-data to build a user interface that provides menus, keyword search, or both simultaneously.

This approach is available anytime the information to be searched has metadata. For example, if a collection of HTML pages or PDFs has authors, dates, subjects or categories, this information can be used to build menus for browsing. If a product catalog has product characteristics like color, size or price, these also can be used to help the user navigate intelligently.

The ability to select items in a database using attributes is called parametric search, because items are selected using parameters.


Consider a product catalog for clothing. A user could search for products containing the word “shirt” in the title, where color=”Blue”, and price < 20.00, sorted by popularity. The system would return results instantly, and would optionally bring back items containing the words "shirts" (plural), "chemise" and "pullover". If you happened to spell it "shert", it would ask, "Did you mean shirt?" Along with the basic search results, the system could display summarized metadata on all shirts. It could also show all of the categories in which such shirts appeared, and give the number of hits in each category. If the shirts also came in other colors, it could show a menu of colors for the user to select. Suppose further that the items in the catalog all had different sets of characteristics. In a catalog of tools, for example, screwdrivers have a length, weight, and head type; drills have an RPM, torque, and chuck type; and there are documents that have titles, authors, and dates. In a single operation Dieselpoint Search could perform both full-text and parametric search, find appropriate items, and display them even if there are apples and oranges in the same result set. By combining full-text search with database-like search on both structured and unstructured data, Dieselpoint search can offer your users much better search results and a much better experience. Go to the Demos page to see what Search and Navigation™ looks like in action.