Search Relevance Algorithm for Weather Effects Products.

摘要

This paper is concerned with providing the user with an efficient way to find information, specifically weather effects products within a Service Oriented Architecture. The work outlined in this paper pertains to searching and ranking weather effects products from the EVIS (Environmental Visualization) data provider. EVIS was a data provider to a Federated Search engine in the NCES (Network Centric Enterprise Services) ECB (Early Capabilities Baseline). Several off the shelf search solutions are examined and a custom search/relevance algorithm is discussed. This algorithm is based on the idea that searching weather products is more akin to a database search. The paper concludes with at a look at cross provider relevance and the complications that arise with a larger scale, growing SOA.