Konfigurasi Backpropagation Neural Network Ringan untuk Klasifikasi Kesuburan Tanah Berbasis Nilai NPK: Analisis Hyperparameter dan Uji Derau Simulasi

Marzuarman Marzuarman; Agustiawan Agustiawan; Mus Ali Alsabah; Azizul Azizul; M. Afridon; Jefri Lianda; Fandy Octafiandri; Cindy Witra Nuansa

doi:10.59061/jentik.v4i3.1622

Authors

Marzuarman Marzuarman Politeknik Negeri Bengkalis
Agustiawan Agustiawan Politeknik Negeri Bengkalis
Mus Ali Alsabah Politeknik Negeri Bengkalis
Azizul Azizul Politeknik Negeri Bengkalis
M. Afridon Politeknik Negeri Bengkalis
Jefri Lianda Politeknik Negeri Bengkalis
Fandy Octafiandri Politeknik Negeri Bengkalis
Cindy Witra Nuansa Politeknik Negeri Bengkalis

DOI:

https://doi.org/10.59061/jentik.v4i3.1622

Keywords:

Backpropagation Neural Network, Lightweight Configuration, Noise Testing, NPK Sensor, Soil Fertility

Abstract

Soil fertility classification from nitrogen (N), phosphorus (P), and potassium (K) values can be embedded in low-cost devices, but the architecture and hyperparameters of a Backpropagation Neural Network (BPNN) are often chosen by trial and error, and the effect of sensor uncertainty is rarely examined. This study aims to find a lightweight BPNN configuration and evaluate it under simulated sensor noise. A dataset of 200 soil samples from Bengkalis Island and Sungai Pakning, labeled into three fertility classes, was used. Hidden-layer size, learning rate, and number of epochs were explored by grid search with 5-fold stratified cross-validation repeated three times, and Gaussian noise of 0–20% of each feature's range was added to the test data. Networks with at least two hidden neurons reached 100.00% accuracy without noise, whereas one hidden neuron reached 94.50%. Within 25 epochs, accuracy rose from 32.00% at a learning rate of 0.0001 to 100.00% at 0.1. The lightweight 3–2–3 configuration (17 parameters) matched the 3–12–8–3 reference (179 parameters) at every noise level, with 95.87% versus 95.57% accuracy at 10% noise, and trained about 30 times faster. A two-neuron BPNN is therefore sufficient for this dataset. The findings are limited to simulated noise and 200 samples, so validation with real RS485 sensor readings is required.

References

Ainurohman, Y. C., Fiolana, F. A., & others. (2026). Perancangan Sensor NPK Menggunakan JST Sebagai Pengambil Keputusan. ALINIER: Journal of Artificial Intelligence & Applications, 7(1), 1–12. https://doi.org/https://doi.org/10.36040/alinier.v7i1.16149

Bergstra, J., & Bengio, Y. (2012). Random search for hyper-parameter optimization. Journal of Machine Learning Research, 13, 281–305.

Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press.

Hastie, T., Tibshirani, R., & Friedman, J. (2009). The Elements of Statistical Learning: Data Mining, Inference, and Prediction (2nd ed.). Springer.

Haykin, S. (2009). Neural Networks and Learning Machines (3rd ed.). Pearson.

Islam, M. R., Oliullah, K., Kabir, M. M., Alom, M., & Mridha, M. F. (2023). Machine learning enabled IoT system for soil nutrients monitoring and crop recommendation. Journal of Agriculture and Food Research, 14, 100880. https://doi.org/10.1016/j.jafr.2023.100880

Kingma, D. P., & Ba, J. (2015). Adam: A Method for Stochastic Optimization. 3rd International Conference on Learning Representations (ICLR 2015).

Kohavi, R. (1995). A study of cross-validation and bootstrap for accuracy estimation and model selection. Proceedings of the 14th International Joint Conference on Artificial Intelligence, 2, 1137–1143.

LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521, 436–444.

Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., & Duchesnay, E. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12, 2825–2830.

Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986). Learning representations by back-propagating errors. Nature, 323, 533–536. https://doi.org/10.1038/323533a0

Saleem, S., Mushtaq, N. U., Rasool, A., Shah, W. H., Tahir, I., & Rehman, R. U. (2023). Plant nutrition and soil fertility: Physiological and molecular avenues for crop improvement. In Sustainable Plant Nutrition: Molecular Interventions and Advancements for Crop Improvement (pp. 23–49). Academic Press. https://doi.org/10.1016/B978-0-443-18675-2.00009-2

Samanta, R., & Saha, B. (2026). Affordable precision agriculture: A deployment-oriented review of low-cost, low-power Edge AI and TinyML for resource-constrained farming systems. ArXiv Preprint ArXiv:2603.15085.

Sokolova, M., & Lapalme, G. (2009). A systematic analysis of performance measures for classification tasks. Information Processing & Management, 45(4), 427–437.

Sujatha, M., & Jaidhar, C. D. (2021). Classification of soil fertility using machine learning-based classifier. International Conference on Secure Cyber Computing and Communications (ICSCCC 2021), 138–143. https://doi.org/10.1109/ICSCCC51823.2021.9478169

Sujatha, M., & Jaidhar, C. D. (2024). Machine learning-based approaches to enhance the soil fertility---A review. Expert Systems with Applications, 240, 122557.

Sujatha, M., Jaidhar, C. D., & Lingappa, M. (2023). 1D convolutional neural networks-based soil fertility classification and fertilizer prescription. Ecological Informatics, 78, 102295.

Sulaeman, Y., Sutanto, E., Kasno, A., Sunandar, N., & Purwaningrahayu, R. D. (2024). Developing and testing a portable soil nutrient detector in irrigated and rainfed paddy soils from Java, Indonesia. Computers, 13(8), 209. https://doi.org/10.3390/computers13080209

Yahaya, S. M., Mahmud, A. A., Abdullahi, M., & Haruna, A. (2023). Recent advances in the chemistry of nitrogen, phosphorus and potassium as fertilizers in soil: A review. Pedosphere, 33(3), 385–406. https://doi.org/10.1016/j.pedsph.2022.07.012