An agent-based framework for intelligent geocoding

Geocoding is essential to translating a physical property address such as for a house, business or landmark into spatial coordinates. These coordinates represent geographic location which is an essential ingredient for location-based services and web mapping. Despite progress in the field of geocoding, there remain a sizable proportion of addresses that are difficult to geocode. The purpose of this research was to explore how agent-based processing, which utilizes the belief, desire, intention (BDI) model, can add intelligence to the geocoding process. The event-driven nature of agent-based processing makes it suitable for use in a web service platform. The event-driven and reactive behaviour is complemented by the ability for goal-directed and non-deterministic behaviour. Overall control of the geocoding process is based on the interactions between agents that represent the geographic elements of an address. Each of these agents operates in parallel, pursuing tasks associated with correcting and preparing their individual address element for geocoding. This results in a geocoding process that has multiple foci of control and is iterative. The same geographic relationships that exist between the address elements also exist between the agents. The agents, as geographic elements, communicate via messages that have content relating to their real-world geographic relationships. These relationships form the basis of the inherent semantics in the intelligent framework. An agentbased prototype was developed for intelligent geocoding. Results indicate that intelligence in geocoding is a product of both context and semantics, at a conceptual level, and control and knowledge, at an implementation level.