Genong
Yu, Indiana State University, gyu@indstate.edu
(Presenting)
Ryan
R
Jensen, Indiana State University, r-jensen@indstate.edu
Paul
W
Mausel, Indiana State University, gemause@isugw.indstate.edu
Vijay
Lulla, Indiana State University, vijay_lulla@mama.indstate.edu
Eduardo
Sonnewend
Brondizio, Indiana University, ebrondiz@indiana.edu
Emilio
F.
Moran, Indiana University, moran@indiana.edu
A distributed spatial information system, the Amazon Information System (AIS), was developed under a NASA funded LBA project to distribute a variety of spatial data associated with Amazon ecosystem studies. Built with an object-oriented approach under a .NET framework, the previous release of the AIS completed a three-tier architecture that promises easy versioning management, program-reuse and sharing, distribution over the Internet, and extension support. Based on previous efforts, , the following objectives were pursued in the latest AIS version: 1) Expansion of current analysis methods in the client software. Specifically, the work focused on the implementation of intelligence-based access and utilization of data, including multiple layered perceptron, Fuzzy ARTMAP, maximum likelihood estimate, maximum likelihood classifier, multiple-response linear/logistic regression, and neural linear feature extractors; 2) Full exploitation of the flexibility and scalability of the client software is made possible by the three-tier architecture. Given the wide range of user groups that the system targets, the AIS needs to tailor its client side functions to cater to different assortments of interests. With the adoption of spatial standards, including GML (geographical markup language) and SVG (scalar vector graphics), the latest phase of AIS development focuses on the implementation of a smart client that combines the benefits of a thick client model where most analyses are completed on the client side with the thin client manageability where most functions are carried out on the server side.
Keywords: smart client, GML, SVG, .NET, GIS