Machine Learning-Based Classification of Gnetum gnemon Fiber Tensile Behavior Using the K-Nearest Neighbors Algorithm

S. Suryadi; Nicodemus Rahanra; Hermanus Jemy Suripatty; Wardhana Wahyu Dharsono

doi:10.35877/454RI.jinav4949

PDF Get rights and content

JINAV: Journal of Information and Visualization

Vol. 7 No. 2 (2026), Pages 257-269 · 2026-08-31

OriginalPaper | Open access | Published: August 31, 2026

Machine Learning-Based Classification of Gnetum gnemon Fiber Tensile Behavior Using the K-Nearest Neighbors Algorithm

S. Suryadi, Nicodemus Rahanra, Hermanus Jemy Suripatty, Wardhana Wahyu Dharsono

JINAV: Journal of Information and Visualization, Vol. 7 No. 2 (2026), pp. 257-269 https://doi.org/10.35877/454RI.jinav4949 Published: 2026-08-31

Get Permissions

Abstract

Natural fibers have been widely developed as reinforcement materials for composite structures due to their favorable mechanical properties, low density, and environmental sustainability. One promising natural fiber is Gnetum gnemon fiber. The mechanical characteristics of fibers are commonly obtained through tensile testing, which generates data representing the relationship between displacement and force. Manual analysis of large volumes of tensile test data is time-consuming and may lead to inconsistencies in data interpretation. This study aims to classify the tensile behavior of Gnetum gnemon fiber using the K-Nearest Neighbors (KNN) machine learning algorithm based on displacement and force parameters. The dataset consisted of 1,561 observations distributed across nine sample classes. Prior to classification, the data underwent preprocessing, including dataset restructuring, data cleaning, Min-Max normalization, and splitting into 80% training data and 20% testing data. The KNN model was developed using the Euclidean distance metric to determine class membership based on the nearest neighbors and was evaluated using accuracy, precision, recall, F1-score, and a confusion matrix. The experimental results showed that the KNN model achieved an accuracy of 69.4% with an average F1-score of approximately 70%. The prediction visualization and confusion matrix indicated that most samples were successfully classified into their respective classes, although some misclassifications occurred among classes with similar mechanical characteristics. These findings demonstrate that the KNN algorithm is capable of classifying the tensile behavior of Gnetum gnemon fiber with satisfactory performance and has the potential to support more efficient analysis of the mechanical characteristics of natural fiber materials.

Keywords

Gnetum gnemon machine learning K-Nearest Neighbors

References (19)

Arunachalam, S. J., Saravanan, R., Sathish, T., & Vinuja, G. (2026). Effect of water absorption on the behaviour of jute/kenaf fiber at various lengths by ANN and RSM modelling. Next Materials, 11, 101726.
Beldar, P., & Kadbhane, S. (2024). Utilizing machine learning to forecast mechanical characteristics of NaOH-treated jute fiber reinforced composite materials. Materials Letters, 377, 137411.
Covallane, K., Johan, D., Singh, R. K., Sinha, R., Boodala, D., Jaiswal, K. K., & Kumar, K. S. (2025). Experimental and Numerical Investigation of Jute Fibre-Reinforced Composite, a Sustainable Material for Green Energy. Engineering Proceedings, 95(1), 17.
Ganesan, V., Shanmugam, V., Kaliyamoorthy, B., Sanjeevi, S., Shanmugam, S. K., Alagumalai, V., Krishnamoorthy, Y., Försth, M., Sas, G., & Javad Razavi, S. M. (2021). Optimisation of mechanical properties in saw-dust/woven-jute fibre/polyester structural composites under liquid nitrogen environment using response surface methodology. Polymers, 13(15), 2471.
Jayakrishna, K., Rajesh, M., Soundhar, A., Surendhar, K., & Sultan, M. T. H. (2018). Modeling and predicting the mechanical properties of nano clay dispersed natural fiber hybrid intra-ply woven fabric composite by using response surface methodology. In Advanced Manufacturing and Materials Science: Selected Extended Papers of ICAMMS 2018 (pp. 221–233). Springer.
Kanthavel, K. (2022). Optimizing various parameters influencing mechanical properties of banana / coir natural fiber composites using grey relational analysis and artificial neural network models. https://doi.org/10.1177/1528083720930304
Mahesh, V., Mahesh, V., & Harursampath, D. (2022). Physio-mechanical characterization of jute/kevlar hybrid composites coupled with MADM approach for selection of composites. Journal of Natural Fibers, 19(15), 11105–11113.
Murugan, A., Barik, D., Faisal, R. S., & Mani, M. (2025). Process parameter optimization for enhanced mechanical and thermal properties of kenaf / jute hybrid composites using grey fuzzy logic. 1–25.
Rahman, M., Ismail, A. E., Ramli, M. F., Hani, A., & Rashid, A. (2026). Prediction of tensile and flexural properties of fiber-reinforced epoxy composites using machine learning models Prediction of tensile and flexural properties of fiber-reinforced epoxy composites using machine learning models.
Saaidia, A., Belaadi, A., Boumaaza, M., & Alshahrani, H. (2023). Effect of Water Absorption on the Behavior of Jute and Sisal Fiber Biocomposites at Different Lengths?: ANN and RSM Modeling Effect of Water Absorption on the Behavior of Jute and Sisal Fiber Biocomposites at Different Lengths?: ANN and RSM Modeling. Journal of Natural Fibers, 20(1). https://doi.org/10.1080/15440478.2022.2140326
Saaidia, A., Belaadi, A., Boumaaza, M., Alshahrani, H., & Bourchak, M. (2023). Effect of water absorption on the behavior of jute and sisal fiber biocomposites at different lengths: ANN and RSM modeling. Journal of Natural Fibers, 20(1), 2140326.
Sahare, S. B., Jadhav, V. D., Satishkumar, P., Vijayan, D. S., & Saminathan, R. (2024). Performance analysis of Jute/Kenaf fiber polyester hybrid composites for non-structural applications. Interactions, 245(1), 164.
Selvaraj, N., Velusamy, S., Tharmalingam, R., & Sivakumar, V. (2026). Experimental Characterization and AI-Based Prediction of Physical and Thermal Properties of Hybrid Fiber-Reinforced Polymer Matrix Composites Using Deep Learning Framework. Fibers and Polymers, 1–29.
Sri Sabarinathan, R., Yazerarabath, J., Thendralvan, S., Srimathi, R., Srileka, A., & Suvathika, C. (2026). Prediction and optimization of mechanical and moisture properties of jute–coir hybrid composites using a DBN–KNN–GWO model. Textile Research Journal, 00405175261458660.
Sundaravadivel, P., & Satheeskumar, S. (2026). Advanced Machine Learning Assisted Prediction of Multi-Fiber Hybrid Natural Fiber/Epoxy Composites Based on Jute, Coconut Coir, Banana, and Pineapple Leaf Fibers. Journal of Materials Research and Technology.
Swain, P. T. R., & Biswas, S. (2018). A comparative analysis of physico?mechanical, water absorption, and morphological behaviour of surface modified woven jute fiber composites. Polymer Composites, 39(8), 2952–2960.
Turyahabwe, A., Dennison, M. S., & George, O. S. (2026). Investigation of multi-performance optimization of banana/bark cloth reinforced epoxy composites using grey relational analysis for automotive interior applications. Scientific Reports, 16(1), 14615.
Upadhyay, V., Kumar, R., & Sharma, C. (2025). Mechanical Properties and Water Absorption Behavior of Jute/Nettle Hybrid Epoxy Composite. In Advanced Functional and Composite Materials (pp. 167–177). CRC Press.
Zhang, W., & He, C. (2022). Investigation and prediction of mechanical properties of hot water treated jute/poly (lactic acid) composite laminates using response surface methodology and genetic algorithm. Polymer Composites, 43(10), 7170–7186.

How to Cite

Suryadi, S., Rahanra, N., Suripatty, H. J., & Dharsono, W. W. (2026). Machine Learning-Based Classification of Gnetum gnemon Fiber Tensile Behavior Using the K-Nearest Neighbors Algorithm. JINAV: Journal of Information and Visualization, 7(2), 257–269. https://doi.org/10.35877/454RI.jinav4949