TY - GEN
T1 - Intelligent Edge-IoT Platform for Corn Yield Prediction
AU - Peñaloza Julio, Carlos Jesus
AU - Aranda Lopez King, Juan Manuel
AU - Tello Oquendo, Luis Patricio
AU - Astudillo Salinas, Fabián
AU - Colcha-Ortiz, Raquel
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
PY - 2026/2/8
Y1 - 2026/2/8
N2 - This study aims to contribute to the efficient use of natural resources in corn cultivation; this includes the precise control of fertilizers and, in general, the monitoring of a crop’s variables to ensure an improvement in production. We introduce a technological and integral solution based on Wireless Sensor Networks (WSNs), Machine Learning (ML), and Edge Computing (EC) to monitor the behavior of a corn crop. At present, the statistics of the entities that monitor agricultural production indicators and the information obtained in the field are inconsistent. The farmer applies eight (8) times less fertilizer than the national average, resulting in twice the production of a non-technified crop and 1.5 times more production if there is some type of technification. Using the proposed platform, which comprises WSN, ML, and EC, contributes to food security by strengthening sustainable agriculture; this aligns with the goal of positively impacting agricultural productivity and the income of small producers. We developed a supervised learning model using Naive Bayes, which was found to have 87% sensitivity and 73% specificity, although the accuracy is 60%.
AB - This study aims to contribute to the efficient use of natural resources in corn cultivation; this includes the precise control of fertilizers and, in general, the monitoring of a crop’s variables to ensure an improvement in production. We introduce a technological and integral solution based on Wireless Sensor Networks (WSNs), Machine Learning (ML), and Edge Computing (EC) to monitor the behavior of a corn crop. At present, the statistics of the entities that monitor agricultural production indicators and the information obtained in the field are inconsistent. The farmer applies eight (8) times less fertilizer than the national average, resulting in twice the production of a non-technified crop and 1.5 times more production if there is some type of technification. Using the proposed platform, which comprises WSN, ML, and EC, contributes to food security by strengthening sustainable agriculture; this aligns with the goal of positively impacting agricultural productivity and the income of small producers. We developed a supervised learning model using Naive Bayes, which was found to have 87% sensitivity and 73% specificity, although the accuracy is 60%.
UR - https://link.springer.com/chapter/10.1007/978-3-031-98768-7_47
UR - https://www.scopus.com/pages/publications/105030269934
U2 - 10.1007/978-3-031-98768-7_47
DO - 10.1007/978-3-031-98768-7_47
M3 - Proceedings
SN - 978-3-031-98767-0
VL - 1516
T3 - Lecture Notes in Networks and Systems
SP - 805
EP - 821
BT - Proceedings of the International Conference on Computer Science, Electronics and Industrial Engineering, CSEI 2024 - Volume 1
A2 - Garcia, Marcelo V.
A2 - Reyes, John-Paul
A2 - Nuñez, Carlos
A2 - Gordón-Gallegos, Carlos
PB - Springer
ER -