Data mining can empower grower's crop decision making
School of Computer and Information Science, Edith Cowan University
Place of Publication
Perth, Western Australia
School of Computer and Security Science
The success of Western Australia crop growers is largely dependent upon the crop variety choices made by them. These decisions are made based on recommendations from government and private organisations. The three main factors that can afftect these decisions and the subsequent outcome are the quality of information, its presentation and interpretation by the grower. Currently, research and crop variety trial data is collated, statistically analysed and presented to growers as a crop variety sowing guide in an annual publication. These publications include recommendations on th evarieties best suited to particular growing environments. The main concern of this process is the interpretation of these recommendations by individual growers. It is evident that growers need to be educated in ways to interpret and use such information in the context of their farming system. Data mining may be one approach that can be used to address this concern as well as to other factors that affect grower decision on crop variety. This research aimed to identify whether data mining would be an effective process to improve Western Australia wheat grower's decision-making on crop variety choices. The research used a case study approach. A data set of crop research trials conducted from 1980 to 2005 was used in the study. This data set contained crop variety performance information from various cropping environments within Western Australia. Various data mining cluster techniques were used to identify unique cropping environments within Western Australia. Results from this process revealed that these unique cropping environments did not match the currently used predefined growing zones. These results would suggest that current information provided to growers on crop variety...
This document is currently not available here.