SavedAdd your journal

Journals / Journal of Artificial Intelligence and Data Science / 2026 / Cilt: 6 Sayı: 1

Comparative Analysis of Fabric Composition and Backing Material Effects on Textile Pressure Sensors Using Random Forest Classifier

Rastgele Orman Sınıflandırıcı Kullanılarak Kumaş Bileşimi ve Destek Malzemesinin Tekstil Basınç Sensörleri Üzerindeki Etkilerinin Karşılaştırmalı Analizi

Gözde Çay(İZMİR KATİP ÇELEBİ ÜNİVERSİTESİ)0000-0003-1985-1772
Pages
56–63
Publication Date
2026-06-29
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The results showed that the Random Forest model successfully differentiated signal windows originating from different textile conditions (accuracy: 91.8% and 95.7% respectively).

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Textile-based sensors are promising candidates for wearable health monitoring systems due to their flexibility, comfort, and ease of integration into garments.
📊
Key Findings & Data:However, their performance can be significantly influenced by the types of fabric and backing material used during fabrication process.
💡
Conclusion & Impact:The results showed that the Random Forest model successfully differentiated signal windows originating from different textile conditions (accuracy: 91.8% and 95.7% respectively).
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:sınıflandırma · textil basınç sensörleri · makine öğrenmesi

Abstract

Textile-based sensors are promising candidates for wearable health monitoring systems due to their flexibility, comfort, and ease of integration into garments. However, their performance can be significantly influenced by the types of fabric and backing material used during fabrication process. This study investigates the effects of different fabric compositions and backing materials using machine learning models. Four fabric types (100% cotton, 94% cotton–6% spandex, 100% nylon, and 82% nylon–18% spandex) were each embroidered with four types of backing materials (tear-away, cut-away, wash-away, and no backing), resulting in a total of sixteen sensors. Each of the sixteen sensors underwent a linearity test in which 0 oz, 2 oz, and 4 oz loads were repeatedly applied and released (~1-second contact) over a continuous 5-minute period for each load (15 minutes per sensor). Then the pressure level and statistical features were extracted from the raw signal for each sensor, and a Random Forest classifier were trained with these features. The results showed that the Random Forest model successfully differentiated signal windows originating from different textile conditions (accuracy: 91.8% and 95.7% respectively). These findings highlight that machine learning models could be used as a tool for sensor characterization by uncovering the hidden relationships within the sensor signals.

Özet

Tekstil tabanlı sensörler, esneklikleri, konforları ve giysilere kolay entegrasyonları nedeniyle giyilebilir sağlık izleme sistemleri için umut vadeden adaylardır. Bununla birlikte, performansları üretim sürecinde kullanılan kumaş ve destek malzemesi türlerinden önemli ölçüde etkilenebilir. Bu çalışma, makine öğrenmesi modelleri kullanarak farklı kumaş bileşimlerinin ve destek malzemelerinin etkilerini araştırmaktadır. Dört farklı kumaş türü (%100 pamuk, %94 pamuk-%6 spandeks, %100 naylon ve %82 naylon-%18 spandeks), her biri dört farklı destek malzemesi türüyle (yırtılabilir, kesilebilir, yıkanabilir ve desteksiz) işlenerek toplam on altı sensör elde edilmiştir. On altı sensörün her biri, 0 oz, 2 oz ve 4 oz yüklerin tekrar tekrar uygulanıp bırakıldığı (~1 saniyelik temas) bir doğrusallık testine tabi tutulmuştur (sensör başına 15 dakika, her yük için 5 dakikalık sürekli bir süre boyunca). Daha sonra, her sensör için ham sinyalden basınç seviyesi ve istatistiksel özellikler çıkarılmış ve bu özelliklerle bir Rastgele Orman sınıflandırıcı eğitilmiştir. Sonuçlar, makine öğrenimi modelinin farklı kumaş türleri ve destek malzemeleri arasında doğru bir şekilde ayrım yapabildiğini göstermiştir (doğruluk oranları sırasıyla %91,8 ve %95,7). Bu bulgular, makine öğrenimi modellerinin sensör karakterizasyonu için bir araç olarak kullanılabileceğini vurgulamaktadır.

🔖

Cite This Article

Çay, G. (2026). Comparative Analysis of Fabric Composition and Backing Material Effects on Textile Pressure Sensors Using Random Forest Classifier. Journal of Artificial Intelligence and Data Science, Cilt: 6 Sayı: 1, 56–63. https://ajindex.com/articles/1618286

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (19)

  1. M. Pan et al., “Soft Controllable Carbon Fibre-based Piezoresistive Self-Sensing Actuators,” Actuators 2020, Vol. 9, Page 79, vol. 9, no. 3, p. 79, Aug. 2020, doi: 10.3390/ACT9030079.
  2. R. R. Ruckdashel, N. Khadse, and J. H. Park, “Smart E-Textiles: Overview of Components and Outlook,” Sensors (Basel), vol. 22, no. 16, p. 6055, Aug. 2022, doi: 10.3390/S22166055.
  3. S. A. Behera et al., “Current Trends on Advancement in Smart Textile Device Engineering,” Adv. Sustain. Syst., vol. 8, no. 12, p. 2400344, Dec. 2024, doi: 10.1002/adsu.202400344.
  4. S. Park and S. Jayaraman, “Smart textiles: Wearable electronic systems,” MRS Bull., vol. 28, no. 8, pp. 585–591, 2003, doi: 10.1557/MRS2003.170.
  5. M. Zhou, J. Wu, L. Huang, M. Miao, Y. Cui, and X. He, “Smart textiles: Sustainable, self-powered, portable, and durable wearable textiles for human health monitoring,” Chemical Engineering Journal, vol. 525, p. 170097, Dec. 2025, doi: 10.1016/J.CEJ.2025.170097.
  6. Y. Yang, Y. Chen, Y. Liu, and R. Yin, “Programmable and Scalable Embroidery Textile Resistive Pressure Sensors for Integrated Multifunctional Smart Wearable Systems,” Advanced Fiber Materials 2025 7:2, vol. 7, no. 2, pp. 574–586, Jan. 2025, doi: 10.1007/S42765-024-00506-5.
  7. O. Atalay, W. Richard Kennon, and M. Dawood Husain, “Textile-Based Weft Knitted Strain Sensors: Effect of Fabric Parameters on Sensor Properties,” Sensors 2013, Vol. 13, Pages 11114-11127, vol. 13, no. 8, pp. 11114–11127, Aug. 2013, doi: 10.3390/S130811114.
  8. M. Su, P. Li, X. Liu, D. Wei, and J. Yang, “Textile-Based Flexible Capacitive Pressure Sensors: A Review,” Nanomaterials, vol. 12, no. 9, p. 1495, May 2022, doi: 10.3390/NANO12091495.
  9. J. G. C. Alfaro, “Analysis and Characterization of Embroidered Textile Strain Sensors for Use in Wearable Mechatronic Devices - ProQuest,” The University of Western Ontario, 2023. Accessed: May 11, 2026. [Online]. Available: https://www.proquest.com/docview/3059441104?fromopenview=true&pq-origsite=gscholar&sourcetype=Dissertations%20&%20Theses
  10. J. G. C. Alfaro and A. L. Trejos, “Design and Fabrication of Embroidered Textile Strain Sensors: An Alternative to Stitch-Based Strain Sensors,” Sensors (Basel), vol. 23, no. 3, p. 1503, Feb. 2023, doi: 10.3390/S23031503.
  11. I. I. Shuvo, A. Shah, and C. Dagdeviren, “Electronic Textile Sensors for Decoding Vital Body Signals: State-of-the-Art Review on Characterizations and Recommendations,” Advanced Intelligent Systems, vol. 4, no. 4, p. 2100223, Apr. 2022, doi: 10.1002/AISY.202100223.
  12. J. wen Zhang, Y. Zhang, Y. yuan Li, and P. Wang, “Textile-Based Flexible Pressure Sensors: A Review,” Polymer Reviews, vol. 62, no. 1, pp. 65–94, 2022, doi: 10.1080/15583724.2021.1901737.
  13. X. Xiao, J. Yin, J. Xu, T. Tat, and J. Chen, “Advances in Machine Learning for Wearable Sensors,” ACS Nano, vol. 18, no. 34, p. 22734, Aug. 2024, doi: 10.1021/ACSNANO.4C05851.
  14. A. Hayat, M. D. Fernando, B. P. Bhuyan, and R. Tomar, “Human activity recognition for elderly people using machine and deep learning approaches,” Information, vol. 13, no. 6, p. 275, Jun. 2022, doi: 10.3390/info13060275.
  15. R. Deeptha et al., “Enhancing human activity recognition for the elderly and individuals with disabilities through optimized Internet-of-Things and artificial intelligence integration with advanced neural networks,” Front. Neuroinform., vol. 18, p. 1454583, 2024, doi: 10.3389/fninf.2024.1454583.
  16. H. K. Alkahtani, N. Al-Kahtani, G. P. Mohammed, and R. Marzouk, “An integration of deep learning models for effective classification of human activity patterns in disabled people using gesture analysis,” Scientific Reports 2025 15:1, vol. 15, no. 1, pp. 43671-, Dec. 2025, doi: 10.1038/s41598-025-27450-7.
  17. G. Ciortuz, H. Hozhabr Pour, M. T. Irshad, M. A. Nisar, X. Huang, and S. Fudickar, “Machine learning models for wearable-based human activity recognition: A comparative study,” Neurocomputing, vol. 650, p. 130911, Oct. 2025, doi: 10.1016/J.NEUCOM.2025.130911.
  18. G. Cay et al., “An E-Textile Respiration Sensing System for NICU Monitoring: Design and Validation,” Journal of Signal Processing Systems 2021 94:6, vol. 94, no. 6, pp. 543–557, Jul. 2021, doi: 10.1007/S11265-021-01669-9.
  19. G. Cay, “Design of a Wearable Multi-Sensory Based Chest Belt Integrated with Iot Functions for Respiration Monitoring - ProQuest.” Accessed: May 13, 2026. [Online]. Available: https://www.proquest.com/docview/2672759302?pq-origsite=gscholar&fromopenview=true&sourcetype=Dissertations%20&%20Theses