Hybrid XGBoost and LSTM Model for Added Amount NPK Fertiliser Prediction in Mango Farm

DOI

10.2991/978-94-6239-745-3_20How to use a DOI?

Keywords

Hybrid Model; Predictive Modeling; Fertiliser Optimization; Agricultural Innovation

Abstract

Efficient fertiliser management is critical for optimising crop yields and reducing environmental degradation. In Malaysia, however, optimal fertiliser use remains a challenge. Many farmers rely heavily on personal experience or weather patterns, often neglecting the actual nutrient status of the soil. This lack of data-driven fertiliser application can lead to both under and over-fertilisation, resulting in reduced productivity and increased environmental risks. This study introduces a hybrid predictive model that combines Extreme Gradient Boosting (XGBoost) and Long Short-Term Memory (LSTM) neural networks to estimate the added amount of nitrogen (N), phosphorus (P), and potassium (K) fertilisers needed for Harumanis mango cultivation. The model integrates the robust feature extraction of XGBoost with LSTM’s capability to model temporal nutrient decay using real-time sensor data, including soil pH, moisture, electrical conductivity (EC), temperature, and rainfall. After hyperparameter optimisation, the hybrid model achieved superior predictive accuracy compared to the individual models, with a Mean Squared Error (MSE) of 0.000581, Mean Absolute Error (MAE) of 0.018369, and R² score of 0.9826. These results confirmed that the hybrid approach effectively captured both the spatial and temporal dynamics of soil nutrients. This study contributes to the advancement of precision agriculture by offering a reliable data-driven method for nutrient recommendations throughout the phenological stages of mango farming.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits any noncommercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if you modified the licensed material. You do not have permission under this license to share adapted material derived from this chapter or parts of it.

Download article (PDF)

TY  - CONF
AU  - Erdy Sulino Bin Mohd Muslim Tan
AU  - Marni Azira Binti Markom
AU  - Allan Melvin Andrew
AU  - Imaduddin Helmi Bin Wan Nordin
AU  - Shahriman Bin Abu Bakar
AU  - Arni Munira Markom
AU  - Pubalan Nadaraja
AU  - Norazila Binti Shoib
AU  - Addzrull Hi-Fi Syam Bin Ahmad Jamil
AU  - Mohd Amri Bin Zainol Abidin
AU  - Farah Hanan Binti Mohd Faudzi
PY  - 2026
DA  - 2026/08/24
TI  - Hybrid XGBoost and LSTM Model for Added Amount NPK Fertiliser Prediction in Mango Farm
BT  - Proceedings of the International Conference on Research, Innovation, Sustainability, and Educations (IRISECON 2026)
PB  - Atlantis Press
SP  - 286
EP  - 308
SN  - 3091-4442
UR  - https://doi.org/10.2991/978-94-6239-745-3_20
DO  - 10.2991/978-94-6239-745-3_20
ID  - Tan2026
ER  -