An Intelligent Volatile Substance Time of Use Prediction System Using 1D Convolutional Neural Network Classifier

Abstract

Volatile substances are chemicals such as inhalants, glues, and solvents which are misused by people. In this work, the time of use of volatile substances is determined using seven personality features and five demographic information. 1D CNN is used for classification whether the user has consumed the volatile substances on last day, last week, last month, last year, or has never consumed. The accuracy obtained is 93.76% which is much higher than all previous studies on volatile substance abuse (VSA).