About | Contact Us | Register | Login
ProceedingsSeriesJournalsSearchEAI
Innovations and Interdisciplinary Solutions for Underserved Areas. 6th EAI International Conference, InterSol 2023, Flic en Flac, Mauritius, September 16-17, 2023, Proceedings

Research Article

On the Use of Machine Learning Technique to Appraise Thermal Properties of Novel Earthen Composite for Sustainable Housing in Sub-Saharan Africa

Cite
BibTeX Plain Text
  • @INPROCEEDINGS{10.1007/978-3-031-51849-2_11,
        author={Assia Aboubakar Mahamat and Moussa Mahamat Boukar},
        title={On the Use of Machine Learning Technique to Appraise Thermal Properties of Novel Earthen Composite for Sustainable Housing in Sub-Saharan Africa},
        proceedings={Innovations and Interdisciplinary Solutions for Underserved Areas. 6th EAI International Conference, InterSol 2023, Flic en Flac, Mauritius, September 16-17, 2023, Proceedings},
        proceedings_a={INTERSOL},
        year={2024},
        month={2},
        keywords={decision tree random forest earthen composite thermal conductivity},
        doi={10.1007/978-3-031-51849-2_11}
    }
    
  • Assia Aboubakar Mahamat
    Moussa Mahamat Boukar
    Year: 2024
    On the Use of Machine Learning Technique to Appraise Thermal Properties of Novel Earthen Composite for Sustainable Housing in Sub-Saharan Africa
    INTERSOL
    Springer
    DOI: 10.1007/978-3-031-51849-2_11
Assia Aboubakar Mahamat1,*, Moussa Mahamat Boukar1
  • 1: Nile University of Nigeria
*Contact email: aassia@aust.edu.ng

Abstract

Earthen based bio-composite reinforced with agricultural waste represent a very important alternative for eco-friendly sustainable building materials. In addition, to the environmental-friendly aspect the use of agro-waste plays a major role in waste management primary by reducing the price related to the waste proper disposal. A novel bio-composite was modeled and tested for its thermal properties to enable the comfort to its habitant. The experimental results were used as primary data to test, train and validate two different machine learning algorithms. The two machine learning models used to predict the thermal conductivity are decision tree regressor (DTR) and random forest (RF). Various inputs were used based on their importance/relationship with the predicted output. The machine learning models were compared based on their efficiency/performance via the evaluation metrics R2, RMSE, MSE and MAE. Decision tree displayed R2= −0.26, RMSE = 0.077, MSE = 0.006 and MAE = 0.05 while random forest displayed values R2= −17.7, RMSE = 0.197, MSE = 0.039 and MAE = 0.119. The results corroborate that both RFR and DTR performed poorly during the predictions, thus they are not suitable for similar composite with the selected input variables.

Keywords
decision tree random forest earthen composite thermal conductivity
Published
2024-02-02
Appears in
SpringerLink
http://dx.doi.org/10.1007/978-3-031-51849-2_11
Copyright © 2023–2025 ICST
EBSCOProQuestDBLPDOAJPortico
EAI Logo

About EAI

  • Who We Are
  • Leadership
  • Research Areas
  • Partners
  • Media Center

Community

  • Membership
  • Conference
  • Recognition
  • Sponsor Us

Publish with EAI

  • Publishing
  • Journals
  • Proceedings
  • Books
  • EUDL