CLASSIFICATION OF VOC VAPORS USING MACHINE LEARNING ALGORITHMS


Aksoy S., Özavşar M. , Altındal A.

International Congress on Fundamental and Applied Sciences, Antalya, Turkey, 19 - 21 October 2021, pp.158

  • Publication Type: Conference Paper / Summary Text
  • City: Antalya
  • Country: Turkey
  • Page Numbers: pp.158

Abstract

Detection of volatile organic compound (VOC) vapors, which are known to have carcinogenic effects, is extremely important and necessary in many areas. In this work, the sensing properties of a CoPc thin film at 6 different VOC vapors (methanol, ethanol, butanol, isopropyl alcohol, acetone, and ammonia) concentrations from 50 to 450 ppm are investigated and it is observed that the interaction between the VOC vapors and CoPc surface is not selective enough. Then, it is showed that using machine learning algorithms the present sensor, which is poorly selective, can be transformed into a more efficient one with higher sensing ability. As a feature, 10 seconds of raw response values taken from steady state region is used without any additional feature extraction technique. Among classification algorithms, 97.1% accuracy is obtained by k-nearest neighbor (KNN) algorithm. The considered feature is also compared with classical maximum response value feature. Classification results indicate that the feature based on 10 seconds of raw response values taken from steady state region is much better than that based on classical maximum response value feature.