Previous research has indicated an increasing need for intelligent automated design. The contention of this project is that Artificial Intelligence (A.I.) techniques can be used to mimic the process of map design in cartography. A suitable environment for such a map system is considered. Attention is focused on methods for identifying and resolving conflicts that occur when spatial data are displayed using cartographic techniques. The research attempts to find a suitable mechanism for describing and identifying spatial conflicts and serves to focus attention on exactly what makes good map design. It appears that human judgement of design requires the understanding of the map as a whole and is based on geographical knowledge and an understanding of spatial processes. This is in addition to the knowledge of design and perception of maps. An appropriate method of description enables evaluation and assessment of the graphic. The potential spatial conflicts that can occur in a map, along with possible solutions for resolving those conflicts, are identified. Automated techniques were devised for identifying features in proximity and resolving those clusters by application of cartographic license (localized feature displacement). Following from this the knowledge governing the use of all generalization techniques is identified and explicitly itemized. A suitable taxonomy of rules is investigated and the knowledge implemented in a rule based system called CLARITY. The rules base contains over one hundred rules. The results and evaluation of the implementation, together with suggested further work conclude this project.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:233935 |
| Date | January 1988 |
| Creators | Mackaness, William Alfred |
| Publisher | Kingston University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://eprints.kingston.ac.uk/20517/ |
Page generated in 0.008 seconds