Mining temperature profile data for shire-level crop yield prediction

Document Type

Conference Proceeding


Faculty of Computing, Health and Science


School of Computer and Security Science




Vagh, Y. , & Xiao, J. (2012). Mining temperature profile data for shire-level crop yield prediction. Proceedings of International Conference on Machine Learning and Cybernetics. (pp. 77-83). Xian, China. Available here


This paper is a continuation of the series of qualitative and quantitative investigations carried out for the processing and analysis of geographic land-use data in an agricultural context. The geographic data was made up of crop and cereal production land use profiles. These were linked to previously recorded climatic data from fixed weather stations in Australia that was interpolated using ordinary krigeing to fit a surface grid. In this investigation, the stochastic average monthly temperature profiles for a selected study area were used to determine the effects on crop production. The areas within the study area were spatially scaled to correspond to individual shires within the South West Agricultural region of Western Australia. The temperature was sampled for three selected years of crop production for 2002, 2003 and 2005. The evaluation was carried out using graphical, correlation and data mining regression techniques in order to detect the patterns of crop production. The patterns suggested that crop production can generally be expected to increase with an increase in temperature during the wheat growing season for some shires.