초록 열기/닫기 버튼
The ability to estimate soil quality has great value for agriculture, especially for low-incomeregions with minimal agricultural and financial resources. This prediction provides users withinformation that is useful in determining whether the soil is suitable for a specific crop, such aspotato (Solanum tuberosum). Farmers in Rwanda lack information on soil quality. There arenot enough soil laboratories to perform the requisite measurements of NPK, pH, and organiccarbon, nor are there enough experts to analyze the data and provide farmers with timelyresults. The prime objective of the proposed study is to develop a predictive framework thatcan estimate soil quality for the ideal cultivation of potato (Solanum tuberosum) considering acase study of Rwanda. In this study, bootstrapping is used to augment the small soil dataset,and fuzzy logic is used to label soil data into four classes of soil suitability, with verification ofthe labeling by soil experts. Several machine learning methods are then tested on the labeleddata, resulting in the classification of suitability for the augmented dataset and an assessment oftheir performance as a way to support experts in predicting soil quality. All machine learningmethods applied were viable, with the best performance achieved using an artificial neuralnetwork. The quantified outcome showed that the adoption of a neural-network-based schemehas an average accuracy of 32% in contrast to other learning schemes. However, 70%-80%accuracy was achieved upon the adoption of fuzzy logic.
