Objective To combine model-based predictions with the experience of forecasters and the knowledge of domain experts A computer system, called Likely, has been developed, which includes an automated system for monitoring forecasts with interactive graphical tools. The modelling of beliefs through stochastic simulation applied to graphical models has been studied. The technical approach included an analysis of why past predictions differed from actual performance. New situations were incorporated within the process. A probabilistic reasoning strategy was developed, based on an influence diagram.Technical approach : - expert will specify why past predictions have differed from actual performance; - expert will notify new situations to incorporate within the forecasting process; - early-warning of significant changes. Probabilistic reasoning strategy will be based on an influence diagram. Fields of science engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringcontrol systems Programme(s) FP2-DOSES - Specific multiannual research and development programme (EEC) of statistical expert systems (DOSES), 1989-1992 Topic(s) Data not available Call for proposal Data not available Funding Scheme Data not available Coordinator SCOTTISH AGRICULTURAL STATISTICS SERVICE (SASS) EU contribution No data Address KING'S BUILDINGS EDINBURGH United Kingdom See on map Total cost No data Participants (6) Sort alphabetically Sort by EU Contribution Expand all Collapse all ECONOMIC AND SOCIAL RESEARCH INSTITUTE Ireland EU contribution No data Address BURLINGTON ROAD 4 See on map Total cost No data Heriot-Watt University United Kingdom EU contribution No data Address Riccarton, Lord Balerno Building EH14 4AS EDINBURGH See on map Total cost No data UNIVERSITY OF DUBLIN Ireland EU contribution No data Address See on map Total cost No data Unisys Espana Spain EU contribution No data Address See on map Total cost No data University of the West of England United Kingdom EU contribution No data Address Coldharbour Lane Frenchay BS16 1QY Bristol See on map Total cost No data Universität Dortmund Germany EU contribution No data Address Otto-Hahn-Straße 18 44227 Dortmund See on map Total cost No data
