Comparison of Machine Learning Regression Models for Predicting Thermal Efficiency of Gaba-Gaba (Sago Leaf stalks) as Traditional Building Material

Sherly Asriany; Muhammad Muzni Herbelubun; Asri A. Muhammad

doi:10.48084/etasr.18666

Authors

Sherly Asriany Department of Architecture, Universitas Khairun, Ternate, Indonesia
Muhammad Muzni Herbelubun Department of Mechanical Engineering, Universitas Khairun, Ternate, Indonesia
Asri A. Muhammad Department of Architecture, Universitas Khairun, Ternate, Indonesia

Volume: 16 | Issue: 3 | Pages: 36552-36558 | June 2026 | https://doi.org/10.48084/etasr.18666

Received: 11 March 2026 | Revised: 8 April 2026 | Accepted: 17 April 2026 | Online: 6 June 2026

Corresponding author: Sherly Asriany

Abstract

This study aims to compare Machine Learning (ML) regression models for predicting the thermal efficiency of gaba-gaba (dried sago leaf stalks), a traditional building material commonly used in tropical regions. Although gaba-gaba is known for its natural thermal insulation properties, quantitative studies on modeling its thermal behavior remain limited. In this research, experimental thermal testing was conducted on gaba-gaba samples of varying thicknesses to measure external and internal surface temperatures, as well as indoor and outdoor air temperatures. From these measurements, thermal conductivity and temperature drop values were calculated. The resulting dataset was used to train and evaluate regression models. Three regression algorithms were implemented and compared: Linear Regression (LR), Support Vector Regression (SVR), and Random Forest Regression (RFR). Model performance was assessed using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and the coefficient of determination (R²). The results indicate that LR achieved the best performance, with an MAE of 0.0056, RMSE of 0.0084, and R² of 0.9976, significantly outperforming the SVR and RFR models. These findings suggest that a simple LR approach can offer high accuracy in modeling the thermal efficiency of gaba-gaba, making it an effective and practical tool to support data-driven sustainable building design using local materials in tropical climates.

Keywords:

thermal conductivity, gaba-gaba, Machine Learning (ML), traditional building materials, Linear Regression (LR), tropical climate

References

F. S. Hafez et al., "Energy Efficiency in Sustainable Buildings: A Systematic Review with Taxonomy, Challenges, Motivations, Methodological Aspects, Recommendations, and Pathways for Future Research," Energy Strategy Reviews, vol. 45, Jan. 2023, Art. no. 101013.

C. Vassiliades, N. Kyriakides, I. Pissourios, and I. Vardopoulos, "Sustainable Architectural Innovations for Energy-Efficient Multi-storey Buildings Through a Novel Taxonomy and Evaluation Approach," Strategic Planning for Energy and the Environment, vol. 43, no. 4, pp. 1195–1224, Dec. 2024.

E. T. Maryanto, A. Setyawan, T. S. B. Maria, and W. Astuti, "Analysis of the Performance of Natural Composite Materials Reinforced with Sago Sheath Fibers as an Alternative Material in Overcoming the Effect of Urban Heat Islands on Buildings," Buildings, vol. 13, no. 1, Dec. 2022, Art. no. 18.

S. Samad, M. R. Rahim, R. T. Lopa, and R. Wikantari, "Modelling of rumbia leaf stalk (GABA-GABA) as wall material and its thermal conductivity capability," International Journal of Civil Engineering and Technology, vol. 8, no. 7, pp. 876–885, July 2017.

Ghassan. O. A. AL-Fakih, R. A. Ilyas, M. R. M. Huzaifah, and A. S. El-Shafay, "Recent advances in sago (Metroxylon sagu) fibres, biopolymers, biocomposites, and their prospective applications in industry: A comprehensive review," International Journal of Biological Macromolecules, vol. 269, June 2024, Art. no. 132045.

A. A. Firoozi, A. A. Firoozi, D. O. Oyejobi, S. Avudaiappan, and E. S. Flores, "Emerging trends in sustainable building materials: Technological innovations, enhanced performance, and future directions," Results in Engineering, vol. 24, Dec. 2024, Art. no. 103521.

I. A. L. Hakiki, S. Wijaksono, and Y. Mariana, "Bamboo as local material for hotel interior finishing," IOP Conference Series: Earth and Environmental Science, vol. 794, no. 1, July 2021, Art. no. 012229.

M. F. V. Adier, M. E. P. Sevilla, D. N. R. Valerio, and J. M. C. Ongpeng, "Bamboo as Sustainable Building Materials: A Systematic Review of Properties, Treatment Methods, and Standards," Buildings, vol. 13, no. 10, Sept. 2023, Art. no. 2449.

J. I. Kindangen, O. H. A. Rogi, and P. H. Gosal, "Thermal Performance Evaluation of Buildings with Coconut Leaf Roofs," Civil Engineering and Architecture, vol. 13, no. 2, pp. 965–975, Mar. 2025.

S. Asriany, M. M. Herbelubun, and M. Jamil, "Sound Absorption Properties of Natural Sago Leaf Fibers (Gaba-Gaba)," International Journal of Design & Nature and Ecodynamics, vol. 19, no. 5, pp. 1763–1768, Oct. 2024.

J. Krzywanski, M. Sosnowski, K. Grabowska, A. Zylka, L. Lasek, and A. Kijo-Kleczkowska, "Advanced Computational Methods for Modeling, Prediction and Optimization—A Review," Materials, vol. 17, no. 14, July 2024, Art. no. 3521.

D. Kömürcü and E. Edis, "Machine Learning Modeling for Building Energy Performance Prediction Based on Simulation Data: A Systematic Review of the Processes, Performances, and Correlation of Process-Related Variables," Buildings, vol. 15, no. 8, Apr. 2025, Art. no. 1301.

Y. Boutahri and A. Tilioua, "Machine learning-based predictive model for thermal comfort and energy optimization in smart buildings," Results in Engineering, vol. 22, June 2024, Art. no. 102148.

R. F. Mustapa, M. A. Hairuddin, A. H. M. Nordin, N. Y. Dahlan, I. M. Yassin, and N. D. K. Ashar, "Machine Learning Baseline Energy Model (MLBEM) to Evaluate Prediction Performances in Building Energy Consumption," Engineering, Technology & Applied Science Research, vol. 14, no. 4, pp. 15938–15946, Aug. 2024.

B. V. Goodarzi and A. R. Bahramian, "Applying machine learning for predicting thermal conductivity coefficient of polymeric aerogels," Journal of Thermal Analysis and Calorimetry, vol. 147, no. 11, pp. 6227–6238, June 2022.

Z. Wu et al., "Using an ensemble machine learning methodology-Bagging to predict occupants’ thermal comfort in buildings," Energy and Buildings, vol. 173, pp. 117–127, Aug. 2018.

N. Bansal, M. Defo, and M. A. Lacasse, "Application of Support Vector Regression to the Prediction of the Long-Term Impacts of Climate Change on the Moisture Performance of Wood Frame and Massive Timber Walls," Buildings, vol. 11, no. 5, Apr. 2021, Art. no. 188.

Y. Chen, Y. Ye, J. Liu, L. Zhang, W. Li, and S. Mohtaram, "Machine Learning Approach to Predict Building Thermal Load Considering Feature Variable Dimensions: An Office Building Case Study," Buildings, vol. 13, no. 2, Jan. 2023, Art. no. 312.

J. Kośny and D. W. Yarbrough, "Thermal Efficiency of Insulation in Building Structures—The Impact of Thermal Bridging," in Thermal Insulation and Radiation Control Technologies for Buildings, J. Kośny and D. W. Yarbrough, Eds. Cham, Switzerland: Springer International Publishing, 2022, pp. 443–486.

T. Kisilewicz, M. Fedorczak-Cisak, and T. Barkanyi, "Active thermal insulation as an element limiting heat loss through external walls," Energy and Buildings, vol. 205, Dec. 2019, Art. no. 109541.

G. Sullivan and M. Spencer, "Heat and temperature," BJA Education, vol. 22, no. 9, pp. 350–356, Sept. 2022.

J. Pebralia, I. Amri, and A. I. Rifa’i, "Measuring convective heat transfer in a room equipped with an air conditioner," Physics Education, vol. 57, no. 5, Aug. 2022, Art. no. 055032.

B. Klemczak, B. Kucharczyk-Brus, A. Sulimowska, and R. Radziewicz-Winnicki, "Historical Evolution and Current Developments in Building Thermal Insulation Materials—A Review," Energies, vol. 17, no. 22, Nov. 2024, Art. no. 5535.

A. F. A. H. Alnuaimi and T. H. K. Albaldawi, "An overview of machine learning classification techniques," BIO Web of Conferences, vol. 97, Apr. 2024, Art. no. 00133.

L. Baresi and G. Quattrocchi, "Training and Serving Machine Learning Models at Scale," in 20th International Conference on Service-Oriented Computing, Seville, Spain, 2022, pp. 669–683.

Y. Zhao, W. Zhang, and X. Liu, "Grid search with a weighted error function: Hyper-parameter optimization for financial time series forecasting," Applied Soft Computing, vol. 154, Mar. 2024, Art. no. 111362.