Projects > FindMe


What do buying a car, selecting a movie, renting an apartment, choosing a restaurant, and picking out a stereo system have in common? All are tasks requiring an individual to choose one item among many, which best meets their unique needs and taste. Find Me systems help users perform such tasks with an assisted browsing approach that combines searching and browsing through large information spaces with knowledge-based assistance.

The aim of assisted browsing is to allow simplified access to information along a multitude of dimensions and from numerous sources. Users become an integral part of the knowledge discovery process, elaborating their information needs in the course of interacting with the system. Knowledge-based retrieval strategies are employed to consider all of the dimensions of the information and present suggestions that lead the user's search in reasonable directions. In a world of ever-expanding information resources, Find Me systems serve as data "tour guides" to assist users in navigating large information spaces.

The Find Me technology allows users to interact with data through examples and tweaks. It allows them to not only say to a system "Get me something like this..." but also refine any suggestions that the system might make by saying "Yes, but a little more..." and have it alter its queries to a database. It gives users a natural method for finding what they want.

At the highest level of abstraction, all Find Me systems are very similar. They contain a database, they retrieve items from it that meet certain constraints, and they rank the retrieved results by some criteria. What gives each Find Me system its character are the details of how this general pattern is instantiated for any given domain, particularly which criteria are used for retrieval and ranking results, which tweaking transformations are incorporated into the system, and which additional knowledge is brought to bear, besides the database itself.

Find Me systems that have been developed include: Car Navigator to help select a new car, Find Me A Movie and Pick a Flick to choose a rental video, Rent Me to find a rental apartment, Entrée to select a restaurant, and Kenwood Home Theater to configure a home theater system. While initially developed for the contexts discussed above, the Find Me approach is applicable to any domain containing a large, fixed set of choices and one sufficiently complex that users would likely be unable to fully articulate their retrieval criteria.