EWA Systems is currently offering open and free usage of its latest e-commerce technologies as web services. One such web service is EWA Systems' Recommendation Server which serves two functions, the personalized recommendation and search of a set of items. This service can easily be integrated on any server- or client-side java enabled web site. For more information or to deploy, contact our Support Team.

Every part of the Recommendation Server is real-time, including the recommendation and search predictions, and the associated learning upon which the prediction models are based. Recommendations are calculated using the full breadth EWA Systems' expert systems and data mining technologies. The accuracy and speed of these recommendations far outstrip that of collaborative filtering, neural network, or any other single technology in terms of prediction accuracy and speed.

EWA Systems incorporated a search function in the recommendation engine since these two functions are so intertwined. The search functions allows searches of all of the item set or subsets of their attributes. The search function finds and corrects misspellings, sound-like, and common typographical errors. The search function also employees an extendable thesaurus, so that words or similar meaning are associated with each other and user's search terms are associated with their finds. All together, this search function, which is straight out of EWA Systems' text mining package, represents the state-of-the-art in search technologies.

The Recommendation Engine can handle a wide array of applications because its item set and user set architecture is extendable, meaning that each item or user can be extended to include almost any data that you have available regarding an item or user. The item and user sets can easily expand into and beyond the millions while still maintaining performance. Finally, the interface through which the Recommendation Engine is accessed is very simple, having just a simple interfaces, and accessed through a simple RMI or web service. All of the details are available in the JavaDocs and User's Guide.