Patent attributes
A flexible, intelligent electronic procurement method and system that emulates and learns from the adaptive behavior of a user trying to find a product in an electronic catalog. In contrast to existing methods and systems, the invention allows for searching heterogeneous catalogs in virtually any format, and does not require pre-defined hierarchies, nor the pre-loading of vendor catalog contents, nor the scrubbing of vendor data. It does not impose fixed structures on the user, it does not require the user to start ‘from scratch’ when looking for a new item, and it does not force the user to think like someone else. Instead, the invention allows the user to choose his own way to navigate catalogs of items, and then, by remembering successful search scenarios, and storing that knowledge in a dynamic, growing database (or ‘forest’) of collected search paths, or ‘find-trees’, the invention grows and improves, that is, it evolves organically over time. As the collection of find-trees can be accessed by other users, the utility of the invention increases over time once implemented in a given environment. Further, the invention is able to do this in a time- and space-efficient manner (that is, without the forest growing too tangled, dense or large) not only because of its method and architecture, but also through the use of autonomous background processes that continually prune and update the tree structures for optimum storage and navigation.