More and more geo-spatial information has been disseminated to the Internet based on WebGIS architecture. Some of
these online mapping applications have already been widely used in recent years, such as Google map, MapQuest,
go2map, mapbar. However, due to the limitation of web map technology and transmit speed of large geo-spatial data
through the Internet, most of these web map systems employ (pyramid-indexed) raster map modeling technology. This
method can shorten server's response time but largely reduces the flexibility and visualization effect of the web map
provided. It will be difficult for them to adaptively change the map contents or map styles for variant user demands. This
paper propose a new system architecture for adaptive web map service by integrating latest network technology and web
map technology, such as SVG, Ajax, user modeling. Its main advantages include: Firstly, it is user customized. In this
proposed map system, user can design the map contents, styles and interfaces online by themselves; secondly, it is more
intelligent. It can record user interactive actions with the system, analyze user profiles, predict user behavior. User's
interests will be obtained and tasks will be suggested based on different user models, which are generated from the
system. For instance, if a new user login in, the nearest user model will be matched and some interactive suggestions will
be provided by the system for the user. It is a more powerful and efficient way for spatial information sharing.
This paper first discusses the main system architecture of adaptive spatial information service which consists of three
parts: user layer, map application layer and database layer. User layer is distributed on client side which includes Web
map (SVG) browser, map renderer and map visualization component. Application layer includes map application server,
user interface generation, user analysis and user modeling, etc. Based on user models, map content, style and user
interactive interface can be predicted and created to fit the user's interests. The database layer includes multi sources of
database, such as thematic spatial database, User Generated Content (UGC) database and others. Secondly, the detailed
technologies of main system components are discussed by the paper, such as user modeling, personal map generation,
etc. Thirdly, adaptive map example is designed and implemented to illustrate the feasibility of the technical solution.
|