Identifying key crop performance traits using data mining

A range of crop related information is distributed to farmers by public and private breeding organizations as well as seed merchants. Farmers use this information to make critical cropping choices which can improve their farm profitability. Frequently, this information highlights the advantages of a particular variety over other released varieties. However, such information is generic and may not be applicable for all farming situations. Better information processes exist that can improve the quality and reliability of this information for individual farming situations. The application of data mining techniques to crop research data enables the customization of information to each individual grower 19s farming situation. The challenge from a research perspective is to identify the key attributes that determine crop performance across different farming situations such as geographic location, soil types, and seasonal conditions. The key attributes include nutrition and soil type, grain yield and quality, sowing and harvest dates and tolerance to environmental stresses. This paper applies data mining techniques to explain this crop performance variability. The results from which can be used by growers to identify particular combination of traits should be used to identify high performing varieties. This study used several techniques to identify differences in crop performance across the different geographic regions. Our research findings suggest that growers could use such data mining techniques to identify high performing varieties for their specific locations and farming practices through the adoption of particular varieties