A review of spatial expert systems: Do they still have a role to play?

Abstract

Expert systems (ES) are one of the oldest cognitive technologies utilized for capturing expert knowledge and emulating decision making. They have been used for solving ill-structured or semi-structured problems from various fields including the geospatial domain. After a growth period of spatial ES in the 1990s followed by slower growth in the 2000s, the decline that has occurred during the last seven years is dramatic, raising a twofold question: Do they still have a role to play and what is the future of spatial automated reasoning? Both questions need to take into account the current advances in computing technologies such as the Internet of Things (IoT), Big Data, cloud computing, the use of social media, crowdsourcing, the considerable increase in computer power available and the revolution of cognitive computing. The latter is a new era of computing, involving the rise of cognitive platforms such as IBM Watson, which may change the way humans interact with computing systems and make decisions about complex problems. Therefore, ESRI has collaborated with IBM Watson in 2016 to enhance the GIS community and industry by supporting spatial-based decision making. In the light of these considerations, this paper provides a brief overview of ES and spatial ES and tries to answer the aforementioned questions. These facts indicate that the development of integrated spatial based cognitive systems within a cloud-based environment, connected with the IoT and other Big Data sources, which will be embedded within customized problem domain knowledge to automate the reasoning for complex spatial-based problems, is just on the horizon.