An improved decision tree algorithm and its application in maize seed breeding

The application of information technology in agriculture accelerates the digitization of agriculture information. How to analyze mass of agriculture data and reveal inherent knowledge are becoming new research points. Effective application of data mining technology can achieve this purpose. This paper presents a new improved CA algorithm based on traditional decision tree method. It introduces a pretreatment theory about double dimension reduction which can deal with large and high-scale datasets. By using CA algorithm in maize seed breeding, we can analyze the potential rules and find useful information from it to direct the growth of maize. The experiments show the improved CA algorithm can obtain more intuitive and efficient information.