(University of Muhammadiyah Malang, Indonesia)(2) Nisrina Nurhafizhah
(University of Muhammadiyah Malang, Indonesia)(3) Amrul Faruq
(University of Muhammadiyah Malang, Indonesia)(4) * Achmad Fauzan Hery Soegiharto
(University of Muhammadiyah Malang, Indonesia)(5) Muhammad Ilham Perdana
(University of Muhammadiyah Malang, Indonesia)*corresponding author
AbstractAccelerometer-based monitoring has become an important approach in precision livestock farming, but achieving both high sensitivity for rare, health-relevant behaviors and computational efficiency remains challenging. Deep learning methods can address class imbalance but are often unsuitable for edge deployment due to their computational cost. This study evaluates the robustness of a lightweight decision-level fusion framework for classifying imbalanced cattle behavior using triaxial accelerometer data. To ensure rigorous evaluation, the Synthetic Minority Over-sampling Technique (SMOTE) was applied only in the training feature space to prevent data leakage. Because the dataset is strongly imbalanced, with Salt Licking (SLT) as the minority yet health-relevant class, model performance was assessed using Macro-F1 for global robustness and Recall_SLT for rare-event sensitivity. Individual tree-based models achieved strong results, with the best single model, XGBoost, obtaining 93.70% accuracy and 0.9421 Macro-F1. The proposed soft-voting fusion of Extra Trees, XGBoost, and CatBoost further improved performance to 94.21% accuracy and 0.9447 Macro-F1, with a statistically significant gain over the best single model (Wilcoxon signed-rank test, (p=0.0326). The framework also maintained strong minority-class recognition, with SLT achieving precision = 0.9571, recall = 0.9853, and PR-AUC = 0.9990. These results show that lightweight decision-level fusion can improve robustness and rare-event sensitivity without temporal deep learning, making it suitable for resource-constrained edge monitoring in livestock systems.
KeywordsCattle Behavior Classification; Decision Level Fusion; Edge AI; Imbalanced Data; SMOTE; Wearable Sensors
|
DOIhttps://doi.org/10.26555/ijain.v12i2.2451 |
Article metricsAbstract views : 200 | PDF views : 5 |
Cite |
Full Text Download
|
References
[1] D. Lovarelli et al., “Development of a New Wearable 3D Sensor Node and Innovative Open Classification System for Dairy Cows’ Behavior,” Animals, vol. 12, no. 11, p. 1447, Jun. 2022, doi: 10.3390/ani12111447.
[2] K. El Moutaouakil and N. Falih, “A comparative study on time series data-based artificial intelligence approaches for classifying cattle feeding behavior,” Indones. J. Electr. Eng. Comput. Sci., vol. 33, no. 1, p. 324, Jan. 2024, doi: 10.11591/ijeecs.v33.i1.pp324-332.
[3] D. Pavlovic et al., “Classification of Cattle Behaviours Using Neck-Mounted Accelerometer-Equipped Collars and Convolutional Neural Networks,” Sensors, vol. 21, no. 12, p. 4050, Jun. 2021, doi: 10.3390/s21124050.
[4] K. Pongsanun, T. Apirak, L. Aekaluck, C. Sanphet, and I. Chaidate, “Classification of individual dairy cow behaviors using accelerometer, gyroscope, and integrated sensor models,” BMC Vet. Res., vol. 21, no. 1, p. 677, Nov. 2025, doi: 10.1186/s12917-025-05092-1.
[5] Á. Michelena, E. Jove, Ó. Fontenla-Romero, and J.-L. Calvo-Rolle, “Evaluation of One-Class Techniques for Early Estrus Detection on Galician Intensive Dairy Cow Farm Based on Behavioral Data From Activity Collars,” ADCAIJ Adv. Distrib. Comput. Artif. Intell. J., vol. 13, p. e32508, Dec. 2024, doi: 10.14201/adcaij.32508.
[6] D.-N. Tran, V.-M. Do, M.-T. Vi, and D.-T. Tran, “Robust Real-Time Analysis of Cow Behaviors Using Accelerometer Sensors and Decision Trees with Short Data Windows and Misalignment Compensation,” Comput. Mater. Contin., vol. 83, no. 2, pp. 2525–2553, Apr. 2025, doi: 10.32604/cmc.2025.062590.
[7] M. W. Iqbal, I. Draganova, P. C. H. Morel, and S. T. Morris, “Validation of an Accelerometer Sensor-Based Collar for Monitoring Grazing and Rumination Behaviours in Grazing Dairy Cows,” Animals, vol. 11, no. 9, p. 2724, Sep. 2021, doi: 10.3390/ani11092724.
[8] V. Mehlin, S. Schacht, and C. Lanquillon, “Towards energy-efficient Deep Learning: An overview of energy-efficient approaches along the Deep Learning Lifecycle,” pp. 1–18, Feb. 2023, Accessed: Jun. 07, 2026. [Online]. Available: http://arxiv.org/abs/2303.01980.
[9] K. El Moutaouakil and N. Falih, “A design of a smart farm system for cattle monitoring,” Indones. J. Electr. Eng. Comput. Sci., vol. 32, no. 2, p. 857, Nov. 2023, doi: 10.11591/ijeecs.v32.i2.pp857-864.
[10] V. Bloch, L. Frondelius, C. Arcidiacono, M. Mancino, and M. Pastell, “Development and Analysis of a CNN- and Transfer-Learning-Based Classification Model for Automated Dairy Cow Feeding Behavior Recognition from Accelerometer Data,” Sensors, vol. 23, no. 5, p. 2611, Feb. 2023, doi: 10.3390/s23052611.
[11] Z. Yu, Y. Han, L. Cha, S. Chen, Z. Wang, and Y. Zhang, “Design of an intelligent wearable device for real-time cattle health monitoring,” Front. Robot. AI, vol. 11, p. 1441960, Nov. 2024, doi: 10.3389/frobt.2024.1441960.
[12] G. Ye and R. Yu, “Spatiotemporal Mapping of Grazing Livestock Behaviours Using Machine Learning Algorithms,” Sensors, vol. 25, no. 15, p. 4561, Jul. 2025, doi: 10.3390/s25154561.
[13] C. Li et al., “Data Augmentation for Inertial Sensor Data in CNNs for Cattle Behavior Classification,” IEEE Sensors Lett., vol. 5, no. 11, pp. 1–4, Nov. 2021, doi: 10.1109/LSENS.2021.3119056.
[14] H.-I. Liu et al., “Lightweight Deep Learning for Resource-Constrained Environments: A Survey,” ACM Comput. Surv., vol. 56, no. 10, pp. 1–42, Oct. 2024, doi: 10.1145/3657282.
[15] K. Chen, D. Zhang, L. Yao, B. Guo, Z. Yu, and Y. Liu, “Deep Learning for Sensor-based Human Activity Recognition,” ACM Comput. Surv., vol. 54, no. 4, pp. 1–40, May 2022, doi: 10.1145/3447744.
[16] M. M. H. Shuvo, S. K. Islam, J. Cheng, and B. I. Morshed, “Efficient Acceleration of Deep Learning Inference on Resource-Constrained Edge Devices: A Review,” Proc. IEEE, vol. 111, no. 1, pp. 42–91, Jan. 2023, doi: 10.1109/JPROC.2022.3226481.
[17] B. Hoffman et al., “A benchmark for computational analysis of animal behavior, using animal-borne tags,” Mov. Ecol., vol. 12, no. 1, p. 78, Dec. 2024, doi: 10.1186/s40462-024-00511-8.
[18] H. Bragança, J. G. Colonna, H. A. B. F. Oliveira, and E. Souto, “How Validation Methodology Influences Human Activity Recognition Mobile Systems,” Sensors, vol. 22, no. 6, p. 2360, Mar. 2022, doi: 10.3390/s22062360.
[19] D. Pavlovic et al., “Behavioural Classification of Cattle Using Neck-Mounted Accelerometer-Equipped Collars,” Sensors, vol. 22, no. 6, p. 2323, Mar. 2022, doi: 10.3390/s22062323.
[20] L. Schmeling et al., “Training and Validating a Machine Learning Model for the Sensor-Based Monitoring of Lying Behavior in Dairy Cows on Pasture and in the Barn,” Animals, vol. 11, no. 9, p. 2660, Sep. 2021, doi: 10.3390/ani11092660.
[21] S. Mahato, H. Bi, and S. Neethirajan, “Dairy DigiD: a keypoint-based deep learning system for classifying dairy cattle by physiological and reproductive status,” Front. Artif. Intell., vol. 8, p. 1545247, Aug. 2025, doi: 10.3389/frai.2025.1545247.
[22] F. A. Nishat, M. F. Mridha, I. Mahmud, M. Alfarhood, M. Safran, and D. Che, “Enhancing Typhoid Fever Diagnosis Based on Clinical Data Using a Lightweight Machine Learning Metamodel,” Diagnostics, vol. 15, no. 5, p. 562, Feb. 2025, doi: 10.3390/diagnostics15050562.
[23] H. Mansourifar and W. Shi, “Deep Synthetic Minority Over-Sampling Technique,” pp. 1–9, Mar. 2020, Accessed: Jun. 07, 2026. [Online]. Available: http://arxiv.org/abs/2003.09788.
[24] X. Zhang et al., “Machine learning for phase prediction of high entropy carbide ceramics from imbalanced data,” npj Comput. Mater., vol. 12, no. 1, p. 22, Jan. 2026, doi: 10.1038/s41524-025-01873-2.
[25] H. James, C. Rial, J. Giordano, and D. Erickson, “Advancing standardized cattle behavior classification with a random forest model evaluated across diverse datasets,” Front. Anim. Sci., vol. 7, p. 1676504, Jan. 2026, doi: 10.3389/fanim.2026.1676504.
[26] A. Tello, V. Degeler, and A. Lazovik, “Too Good To Be True: accuracy overestimation in (re)current practices for Human Activity Recognition,” in 2024 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops), IEEE, Mar. 2024, pp. 511–517. doi: 10.1109/PerComWorkshops59983.2024.10503465.
[27] A. Rojarath and W. Songpan, “Probability-Weighted Voting Ensemble Learning for Classification Model,” J. Adv. Inf. Technol., vol. 11, no. 4, pp. 217–227, Nov. 2020, doi: 10.12720/jait.11.4.217-227.
[28] F. Tian, J. Wang, B. Xiong, L. Jiang, Z. Song, and F. Li, “Real-Time Behavioral Recognition in Dairy Cows Based on Geomagnetism and Acceleration Information,” IEEE Access, vol. 9, pp. 109497–109509, 2021, doi: 10.1109/ACCESS.2021.3099212.
[29] R. Maurya, S. Mahapatra, and L. Rajput, “A Lightweight Meta-Ensemble Approach for Plant Disease Detection Suitable for IoT-Based Environments,” IEEE Access, vol. 12, pp. 28096–28108, 2024, doi: 10.1109/ACCESS.2024.3367443.
[30] G. Menghani, “Efficient Deep Learning: A Survey on Making Deep Learning Models Smaller, Faster, and Better,” ACM Comput. Surv., vol. 55, no. 12, pp. 1–37, Dec. 2023, doi: 10.1145/3578938.
[31] H. Ito et al., “Japanese Black Beef Cow Behavior Classification Dataset,” Zenodo. Accessed: Jun. 07, 2026. [Online]. Available: https://zenodo.org/records/5849025.
[32] Y. Li et al., “Classification and Analysis of Multiple Cattle Unitary Behaviors and Movements Based on Machine Learning Methods,” Animals, vol. 12, no. 9, p. 1060, Apr. 2022, doi: 10.3390/ani12091060.
[33] T. Mladenova et al., “Evaluation of the Efficiency of Machine Learning Algorithms for Identification of Cattle Behavior Using Accelerometer and Gyroscope Data,” AgriEngineering, vol. 6, no. 3, pp. 2179–2197, Jul. 2024, doi: 10.3390/agriengineering6030128.
[34] L. Ding et al., “Predicting the Feed Intake of Cattle Based on Jaw Movement Using a Triaxial Accelerometer,” Agriculture, vol. 12, no. 7, p. 899, Jun. 2022, doi: 10.3390/agriculture12070899.
[35] M. Fatima, O. Rehman, I. M. H. Rahman, A. Ajmal, and S. J. Park, “Towards Ensemble Feature Selection for Lightweight Intrusion Detection in Resource-Constrained IoT Devices,” Futur. Internet, vol. 16, no. 10, p. 368, Oct. 2024, doi: 10.3390/fi16100368.
[36] T. Shahriar, “Comparative Analysis of Lightweight Deep Learning Models for Memory-Constrained Devices,” pp. 1–22, Sep. 2025, Accessed: Jun. 10, 2026. [Online]. Available: http://arxiv.org/abs/2505.03303.
[37] D. Li et al., “Optimized Machine Learning Models for Predicting Core Body Temperature in Dairy Cows: Enhancing Accuracy and Interpretability for Practical Livestock Management,” Animals, vol. 14, no. 18, p. 2724, Sep. 2024, doi: 10.3390/ani14182724.
[38] S. R. Alve, M. Z. Mahmud, S. Islam, M. A. Chowdhury, and J. Islam, “Resource-Efficient Machine Learning Approaches for Multi-Class Threat Detection in IoT Environments,” in 2025 International Conference on Quantum Photonics, Artificial Intelligence, and Networking (QPAIN), IEEE, Jul. 2025, pp. 1–6. doi: 10.1109/QPAIN66474.2025.11171769.
[39] R. Ma, R. Chen, B. Liang, and X. Li, “A XGBoost-Based Prediction Method for Meat Sheep Transport Stress Using Wearable Photoelectric Sensors and Infrared Thermometry,” Sensors, vol. 24, no. 23, p. 7826, Dec. 2024, doi: 10.3390/s24237826.
[40] S. Bergen et al., “A review of supervised learning methods for classifying animal behavioural states from environmental features,” Methods Ecol. Evol., vol. 14, no. 1, pp. 189–202, Jan. 2023, doi: 10.1111/2041-210X.14019.
[41] K. E. ArunKumar et al., “Predicting dry matter intake in cattle at scale using gradient boosting regression techniques and Gaussian process boosting regression with Shapley additive explanation explainable artificial intelligence, MLflow, and its containerization,” J. Anim. Sci., vol. 103, pp. 1–20, Jan. 2025, doi: 10.1093/jas/skaf041.
[42] S. W. A. Sherazi, J.-W. Bae, and J. Y. Lee, “A soft voting ensemble classifier for early prediction and diagnosis of occurrences of major adverse cardiovascular events for STEMI and NSTEMI during 2-year follow-up in patients with acute coronary syndrome,” PLoS One, vol. 16, no. 6, p. e0249338, Jun. 2021, doi: 10.1371/journal.pone.0249338.
[43] A. Elkelany, R. Ross, and S. Mckeever, “Identifying Subject Bias in WiFi-based Human Activity Recognition Evaluation Methods,” in The 32nd Irish Conference on Artificial Intelligence and Cognitive Science (AICS 2024), 2024, pp. 301–313. Accessed: Jun. 10, 2026. [Online]. Available: https://ceur-ws.org/Vol-3910/aics2024_p78.
[44] L. S. AMBATI, O. El-Gayar, L. S. AMBATI, and O. El-Gayar, “Human Activity Recognition: A Comparison of Machine Learning Approaches,” J. Midwest Assoc. Inf. Syst., vol. 2021, no. 1, p. 4, Mar. 2021, Accessed: Jun. 12, 2026. [Online]. Available: https://aisel.aisnet.org/jmwais/vol2021/iss1/4.
[45] M. Gharagozloo, A. Amrani, K. Wittingstall, A. Hamilton-Wright, and D. Gris, “Machine Learning in Modeling of Mouse Behavior,” Front. Neurosci., vol. 15, p. 700253, Sep. 2021, doi: 10.3389/fnins.2021.700253.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
___________________________________________________________
International Journal of Advances in Intelligent Informatics
ISSN 2442-6571 (print) | 2548-3161 (online)
Organized by UAD and ASCEE Computer Society
Published by Universitas Ahmad Dahlan
W: http://ijain.org
E: info@ijain.org (paper handling issues)
andri.pranolo.id@ieee.org (publication issues)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0