Document Type : Research Paper
Authors
1
Biosystems Engineering Dept., Agricultural faculty, Shahid Chamran University of Ahvaz, Ahvaz, Iran,
2
Department of Computer Engineering, Faculty of Electrical and Computer Engineering, Tabriz University, Tabriz, Iran
3
Department of Biosystem Engineering, Faculty of Agriculture, Tabriz University, Tabriz, Iran
4
Biosystems engineering Dept., Faculty of Agriculture, Shahid Chamran University of Ahvaz, Ahvaz, Iran
Abstract
In this research, one of the main goals of Khuzestan sugarcane Agro-industry companies, i.e. increasing the yield of sugarcane fields by using data mining methods, has been investigated. This research is of analytical type and includes the irrigation, drainage, soil and plant data of 1201 farms which were collected from Amirkabir Agriculture Company in 2013 to 2016 crop years. In this research, four algorithms of long-short recurrent neural network (LSTM), multilayer neural network perceptron (MLP), decision tree and support vector machine (SVM) were used, and two dimension reduction methods, principal component analysis (PCA) and algorithm Independent component analysis (ICA) was applied using Python software. In the principal component analysis (PCA), the final variables including crop variety, soil texture, spraying area ratio, soil electrical conductivity, drainage and nitrogen fertilizer were identified. While in the analysis of independent components, the final variables included product variety, soil electrical conductivity, water electrical conductivity, plant age, the number of times of irrigation and soil texture. The results showed that the long-term recurrent neural network (LSTM) algorithm performed better in the principal component analysis (PCA) dimension reduction method. The values of R² equal to 97%, RMSE equal to 51.79, and RRMSE equal to 0.89 were obtained for this algorithm in the PCA method, compared to the ICA method, which had values of R² equal to 91%, RMSE equal to 62.75, and RRMSE equal to 0.798. , which provided better results. This shows that PCA had a better ability to reduce the dimensionality for this model.
Keywords
Main Subjects
)