Predicting Suffocation and Fire Risks Using IoT Systems and Machine Learning Techniques

Öz

This study aims to predict suffocation and fire risks using machine learning (ML) and Internet of Things (IoT) technologies. Suffocation and fire incidents are serious dangers that lead to loss of life and property. While traditional methods may be insufficient in predicting these risks, high-accuracy predictions can be achieved by using ML models and analyzing large data sets obtained with IoT-based sensor data. In this study, a comparative study of ML methods was conducted on data obtained from IoT sensors. The correlation between carbon monoxide (CO), smoke, and liquid petroleum gas (LPG) was used to predict the risk of suffocation, and the correlation between humidity, temperature, and smoke was used to predict the risk of fire. After the pre-processing steps were completed, the data were trained with models developed with linear regression, decision trees, and random forest algorithms. Experimental results show that the decision tree outperformed other algorithms by 99.99%. It was determined that the developed model worked with high accuracy rates and could play an important role in the development of risk management strategies.

Anahtar Kelimeler

• Internet of Things
• Environmental Sensors
• Machine Learning
• Decision Trees
• Risk Prediction

IoT Sistemleri ve Makine Öğrenmesi Teknikleri Kullanılarak Boğulma ve Yangın Risklerinin Tahmini

Öz

Bu çalışmada, makine öğrenmesi (ML) ve Nesnelerin İnterneti (IoT) teknolojilerini kullanarak boğulma ve yangın risklerinin tahmin edilmesi amaçlanmaktadır. Boğulma ve yangın olayları, ciddi, can ve mal kayıplarına yol açan tehlikelerdir. Geleneksel yöntemler bu risklerin tahmin edilmesinde yetersiz kalabilirken, ML modelleri kullanarak ve IoT tabanlı sensör verileriyle elde edilen büyük veri kümelerinin analiziyle yüksek doğrulukta tahminler sağlanabilmektedir. Bu çalışmada, IoT sensörlerinden elde edilen veriler üzerinde ML yöntemlerinin karşılaştırmalı bir incelemesi yapılmıştır. Boğulma riskini tahmin etmek için karbon monoksit (CO), duman ve sıvı petrol gazı (LPG) arasındaki korelasyon, yangın riskini tahmin etmek için ise nem, sıcaklık ve duman arasındaki korelasyon kullanılmıştır. Veriler, ön işleme adımları tamamlandıktan sonra, doğrusal regresyon, karar ağaçları ve rastgele orman algoritmalarıyla geliştirilen modellerle eğitilmiştir. Deneysel sonuçlar, karar ağacının %99,99 doğrulukla diğer algoritmaları geride bıraktığını göstermektedir. Geliştirilen modelin yüksek doğruluk oranı ile çalıştığı ve risk yönetimi stratejilerinin geliştirilmesinde önemli bir rol oynayabileceği belirlenmiştir.

Anahtar Kelimeler

• Nesnelerin İnterneti
• Çevresel Sensörler
• Makine Öğrenmesi
• Karar Ağaçları
• Risk Tahmini

Kaynakça

1. Shaikh M, Ali A, Ahmed R, Shaikh BA. A review on Internet of Things (IoT) based water monitoring system. J Kejuruter 2023; 35(6): 1273-1278.
2. AlZubi AA. IoT-based automated water pollution treatment using machine learning classifiers. Environ Technol 2024; 45(12): 2299-2307.
3. Kaginalkar A, Kumar S, Gargava P, Niyogi D. Review of urban computing in air quality management as smart city service: An integrated IoT, AI, and cloud technology perspective. Urban Clim 2021; 39: 100972.
4. Vo DT, Nguyen XP, Nguyen TD, Hidayat R, Huynh TT, Nguyen DT. A review on the internet of thing (IoT) technologies in controlling ocean environment. Energy Sources Part A: Recovery Util Environ Eff 2021; 43(1): 1-19.
5. Popescu SM, Mansoor S, Wani OA, Kumar SS, Sharma V, Sharma A, Arya VM, ve diğerleri. Artificial intelligence and IoT driven technologies for environmental pollution monitoring and management. Front Environ Sci 2024; 12: 1336088.
6. Khalil K, Elgazzar K, Seliem M, Bayoumi M. Resource discovery techniques in the internet of things: A review. Internet Things 2020; 12: 100293.
7. Al-Obaidi KM, Hossain M, Alduais NAM, Al-Duais HS, Omrany H, Ghaffarianhoseini A. A review of using IoT for energy efficient buildings and cities: A built environment perspective. Energies 2022; 15(16): 5991.
8. Musa AA, Hussaini A, Qian C, Guo Y, Yu W. Open radio access networks for smart IoT systems: State of art and future directions. Futur Internet 2023; 15(12): 380.

9. Li Y, Alqahtani A, Solaiman E, Perera C, Jayaraman PP, Buyya R, Morgan G, Ranjan R. IoT-CANE: A unified knowledge management system for data-centric Internet of Things application systems. J Parallel Distrib Comput 2019; 131: 161-172.
10. Xiang J, Zhao A, Tian GY, Woo W, Liu L, Li H. Prospective RFID sensors for the IoT healthcare system. J Sensors 2022; 2022(1): 8787275.
11. Abdulmalek S, Nasir A, Jabbar WA, Almuhaya MAM, Bairagi AK, Khan MA-M, Kee S-H. IoT-based healthcare-monitoring system towards improving quality of life: A review. Healthcare 2022; 10(10): 1993.
12. Al-Rawashdeh M, Keikhosrokiani P, Belaton B, Alawida M, Zwiri A. IoT adoption and application for smart healthcare: A systematic review. Sensors 2022; 22(14): 5377.
13. Hintaw AJ, Manickam S, Aboalmaaly MF, Karuppayah S. MQTT vulnerabilities, attack vectors and solutions in the Internet of Things (IoT). IETE J Res 2023; 69(6): 3368-3397.
14. Sharma A, Nayyar A, Singh KJ, Kapoor DS, Thakur K, Mahajan S. An IoT-based forest fire detection system: design and testing. Multimed Tools Appl 2023; 83(13): 38685-38710.
15. Riskiawan HY, Gupta N, Setyohadi DPS, Anwar S, Kurniasari AA, Hariono B, Firmansyah MH, Yogiswara Y, ve diğerleri. Artificial intelligence enabled smart monitoring and controlling of IoT-green house. Arab J Sci Eng 2023; 49(3): 3043-3061.
16. Prabu RT, Sarkar M, Chaudhary D, Al Obaid S, Al-ateeq TK, Kalam MA. IoT-enabled groundwater monitoring with k-NN-SVM algorithm for sustainable water management. Acta Geophys 2024; 72(4): 2715-2728.
17. Yan Z, Zhang P, Vasilakos AV. A survey on trust management for Internet of Things. J Netw Comput Appl 2014; 42: 120-134.
18. Olivier F, Carlos G, Florent N. New security architecture for IoT network. Procedia Comput Sci 2015; 52: 1028-1033.
19. Azzedin F, Alhazmi T. Secure data distribution architecture in IoT using MQTT. Appl Sci 2023; 13(4): 2515.
20. Vo DT, Nguyen XP, Nguyen TD, Hidayat R, Huynh TT, Nguyen DT. A review on the internet of thing (IoT) technologies in controlling ocean environment. Energy Sources Part A: Recovery Util Environ Eff 2021; 43(1): 1-19.
21. Stafford G. Environmental Sensor Telemetry Data - Kaggle. Erişim: 04 Mart 2025. [Çevrimiçi]. Erişim adresi: https://www.kaggle.com/datasets/garystafford/environmental-sensor-data-132k/code