On the role of expert systems and numerical taxonomy in soil classification

Expert systems and numerical classification in general are reviewed and their relative ability to improve soil classification systems are discussed. The inductive and deductive elements of expert systems are seen as corresponding to the class establishment and class assignment phases of classification; computerized expert systems have so far been set up only for very simple assignments. To date, numerical classification has been useful in the analysis and organization of local low-level soil data but has been largely untried at the higher global levels of soil classification because of a lack of suitable data and scientific commitment. It is concluded that numerical classification has a potentially useful part to play in establishing soil classes and generating rules for assignment in expert systems. The inference procedures and user interface of expert systems should allow more integration and realistic assignment of available data and increase the usefulness and predictability of soil information, especially if presented in a user-friendly mode.