Evaluating machine learning algorithms for predicting maize yield under conservation agriculture in Eastern and Southern
- PDF / 3,324,483 Bytes
- 14 Pages / 595.276 x 790.866 pts Page_size
- 48 Downloads / 189 Views
Evaluating machine learning algorithms for predicting maize yield under conservation agriculture in Eastern and Southern Africa W. Mupangwa1 · L. Chipindu2 · I. Nyagumbo2 · S. Mkuhlani3 · G. Sisito4 Received: 13 December 2019 / Accepted: 7 April 2020 © Springer Nature Switzerland AG 2020
Abstract Crop simulation models are widely used as research tools to explore the impact of various technologies and compliment field experimentation. Machine learning (ML) approaches have emerged as promising artificial intelligence alternative and complimentary tools to the commonly used crop production models. The study was designed to answer the following questions: (a) Can machine learning techniques predict maize grain yields under conservation agriculture (CA)? (b) How close can ML algorithms predict maize grain yields under CA-based cropping systems in the highlands and lowlands of Eastern and Southern Africa (ESA)? Machine learning algorithms could predict maize grain yields from conventional and CA-based cropping systems under low and high potential conditions of the ESA region. Linear algorithms (LDA and LR) predicted maize yield more closely to the observed yields compared with nonlinear tools (NB, KNN, CART and SVM) under the conditions of the reported study. However, the KNN algorithm was comparable in its yield prediction to the linear tools tested in this study. Overall, the LDA algorithm was the best tool, and SVM was the worst algorithm in maize yield prediction. Evaluating the performance of different ML algorithms using different criteria is critical in order to get a more robust assessment of the tools before their application in the agriculture sector. Keywords Agro-ecology · Big data · Data-driven value creation · Cropping systems · Smallholder agriculture
1 Introduction Crop simulation models are widely used as research tools to explore the impact of various technologies under different biophysical and socio-economic conditions [18, 38]. The use of modelling tools complements field experiments conducted over a short period of time, saves time and financial resources and allows for extrapolation of experimental results to other biophysical and socio-economic conditions [22, 31]. Additionally, simulation models aid in
identifying knowledge gaps, testing hypotheses, designing new experiments and determining the most influential factors in farming systems of interest [21, 40]. Various crop simulation models have been developed, tested and applied for different purposes in agricultural research and development of farming systems. Models widely used to better understand smallholder cropping systems in sub-Saharan Africa (SSA) include, among others, APSIM, DSSAT, Aqua-Crop, Patched-Thirst and SOYGRO [31]. These models have been used to increase
Electronic supplementary material The online version of this article (https://doi.org/10.1007/s42452-020-2711-6) contains supplementary material, which is available to authorized users. * W. Mupangwa, [email protected]; [email protected] | 1International Maize an
Data Loading...
