Prediction of Hygroscopic Properties of Untreated Borassus Fiber Using Ensemble Learning Techniques

Abstract:

—Borassus palm or aethiopium palmyra is a palm species tree, widely spread in sub-Saharan Africa but its fruits don’t have any economic value therefore considered as waste. This study investigated the potential of Borassus fruit fibers (BFF), extracted manually from the underutilized fruit, for various applications by examining their hygroscopic properties. Scanning electron microscopy (SEM) revealed the fibers' unique features, including a relatively large diameter and high affinity for water vapor. A Dynamic Vapor Sorption (DVS) analysis with exposure time varying from 1, 2, 4 until 72h and varied Relative Humidity (from 0 to 90%) with 10% increment was carried out to examine the Sorption-desorption behavior. The characteristic hysteresis behavior of natural fibers was observed, with significant moisture uptake, particularly above 70% RH. The sorption and desorption processes were quantified, revealing a linear relationship between mass change and relative humidity. Furthermore, an Ensemble learning approach, specifically a Gradient Boosting Regression (GBR) model, was developed to predict the hygroscopic behavior of BFF. Trained on the experimental sorption-desorption data, the GBR model demonstrated excellent predictive accuracy, achieving a high R² value of 91.7% and low CV, MSE, and RMSE values (6.9 and 2.6, respectively). These findings highlight the significant influence of relative humidity on BFF moisture content and demonstrate the effectiveness of GBR as a powerful tool for accurately predicting the complex hygroscopic behavior of these fibers. Keywords—Machine Learning techniques, Ensemble Learning, Natural fiber, Hygroscopic properties, sorption/desorption

• Periodical: Materials Science Forum (Volume 1181)
• Pages: 79-85
• DOI: https://doi.org/10.4028/p-T9tURX
• Online since: March 2026
• Authors: Assia Aboubakar Mahamat*, Moussa Mahamat Boukar, Philbert Nshimiyimana
• Keywords: Desorption, Ensemble Learning, Hygroscopic Properties, Machine Learning Techniques, Natural Fiber, Sorption

References

[1] A. A. Mahamat et al., "Dimensionnal stability and strength appraisal of termite hill soil stabilisation using hybrid bio-waste and cement for eco-friendly housing," Heliyon, vol. 8, no. 5, May 2022.

DOI: 10.1016/j.heliyon.2022.e09406

[2] M. Humar et al., "Quality control of thermally modified timber using dynamic vapor sorption (DVS) analysis," Forests, vol. 11, no. 6, Jun. 2020.

DOI: 10.3390/f11060666

[3] P. Sudhakara et al., "Studies on Borassus fruit fiber and its composites with Polypropylene," Composites Research, vol. 26, no. 1, p.48–53, Feb. 2013.

DOI: 10.7234/kscm.2013.26.1.48

[4] A. A. Mahamat et al., "Assessment of hygrothermal and mechanical performance of alkali activated Borassus fiber reinforced earth-based bio-composite," Journal of Building Engineering, vol. 62, Dec. 2022.

DOI: 10.1016/j.jobe.2022.105411

[5] Assia Aboubakar Mahamat, Numfor Linda Bih, and Holmer Savastano Jr, "Optimization of termite mound soil through alkali activation and cement stabilisation for sustainable and eco-friendly construction materials," in International Conference on Multidisciplinary Engineering and Applied Science (ICMEAS), 2021. doi: 978-1-7281-8068-7/20.

DOI: 10.1109/icmeas52683.2021.9692382

[6] W. Wang, Y. Wu, W. Liu, T. Fu, R. Qiu, and S. Wu, "Tensile Performance Mechanism for Bamboo Fiber-Reinforced, Palm Oil-Based Resin Bio-Composites Using Finite Element Simulation and Machine Learning," Polymers (Basel), vol. 15, no. 12, Jun. 2023.

DOI: 10.3390/polym15122633

[7] A. A. Mahamat et al., "A Machine Learning Led Investigation Predicting the Thermos-mechanical Properties of Novel Waste-based Composite in Construction," Waste Biomass Valorization, May 2024.

DOI: 10.1007/s12649-024-02538-9

[8] E. A. Yousef, B. A. Mouhcine, Z. Mounir, and H. A. Adil, "Prediction of compressive strength of self-compacting concrete using four machine learning technics," Mater Today Proc, 2022.

DOI: 10.1016/j.matpr.2022.02.487

[9] Assia Aboubakar Mahamat and Moussa Mahamat Boukar, "Machine learning techniques versus classical statistics in strength predictions of eco-friendly masonry units," in 16th International Conference on Electronics Computer and Computation (ICECCO 2021), Sep. 2021.

DOI: 10.1109/icecco53203.2021.9663760

[10] E. E. Nyakilla et al., "Application of machine learning in the prediction of compressive, and shear bond strengths from the experimental data in oil well cement at 80 °C. Ensemble trees boosting approach," Constr Build Mater, vol. 317, Jan. 2022.

DOI: 10.1016/j.conbuildmat.2021.125778

[11] A. A. Mahamat et al., "Decision Tree Regression vs. Gradient Boosting Regressor Models for the Prediction of Hygroscopic Properties of Borassus Fruit Fiber," Applied Sciences, vol. 14, no. 17, p.7540, Aug. 2024.

DOI: 10.3390/app14177540

[12] A. Megalingam, M. Kumar, B. Sriram, K. Jeevanantham, and P. Ram Vishnu, "Borassus fruit fiber reinforced composite: A review," in Materials Today: Proceedings, Elsevier Ltd, 2021, p.723–728.

DOI: 10.1016/j.matpr.2020.02.750

[13] A. Célino, S. Fréour, F. Jacquemin, and P. Casari, "The hygroscopic behavior of plant fibers: A review," Front Chem, vol. 1, no. JAN, 2014.

DOI: 10.3389/fchem.2013.00043

[14] V. Q. Tran, "Hybrid gradient boosting with meta-heuristic algorithms prediction of unconfined compressive strength of stabilized soil based on initial soil properties, mix design and effective compaction," J Clean Prod, vol. 355, Jun. 2022.

DOI: 10.1016/j.jclepro.2022.131683

[15] A. A. Mahamat and M. M. Boukar, "On the Use of Machine Learning Technique to Appraise Thermal Properties of Novel Earthen Composite for Sustainable Housing in Sub-Saharan Africa," in Innovations and Interdisciplinary Solutions for Underserved Areas, 2023, p.161–170.

DOI: 10.1007/978-3-031-51849-2_11