Premium
Desert: A prototype expert system to advise on land degradation control
Author(s) -
Balachandran C. S.,
Fisher P. F.
Publication year - 1990
Publication title -
land degradation and development
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.403
H-Index - 81
eISSN - 1099-145X
pISSN - 1085-3278
DOI - 10.1002/ldr.3400020104
Subject(s) - land degradation , expert system , vegetation (pathology) , computer science , set (abstract data type) , identification (biology) , component (thermodynamics) , control (management) , environmental resource management , soil retrogression and degradation , desert (philosophy) , environmental science , productivity , land use , agricultural engineering , remote sensing , artificial intelligence , civil engineering , soil water , engineering , geography , soil science , philosophy , pathology , biology , epistemology , programming language , thermodynamics , medicine , botany , physics , macroeconomics , economics
Land degradation is a collection of processes causing the removal of land from agricultural productivity. Four major processes can be identified: soil erosion by water and wind, soil salinization, and vegetation degradation. Much knowledge exists on how these processes may be combated and ameliorated; however, it is in the minds of a relatively few, widely dispersed experts. The authors have developed a prototype computer‐based expert system (DESERT) which provides advice on strategies for control of land degradation. The knowledge was acquired from experts by interviews. The knowledge was then formulated into two components. The first component enables identification of possible technical solutions, while the other imposes socio‐economic and physical constraints on these technical solutions. Thus, a set of acceptable solutions is formulated. The results are presented here as a proof of concept in an attempt to stimulate further development. Later versions of the program will include problem diagnosis components, as well as a mechanism for prioritizing recommendations.