Classifying Different Types of Smokers and Drinkers by Analyzing Body Signals using Machine Learning

Alcohol consumption and smoking are one of the most leading preventable causes of mortality worldwide. Timely detection can play a vital role to prevent these kinds of addictions. To achieve that our study utilizes body signals to develop a machine-learning-based classification system for predicting the types of smokers and drinkers, where we used Logistic Regression, K-Nearest Neighbors, Naïve Bayes, and Random Forest for the classification. It has been found that Naïve Baye and Random Forest both perform well among the four classification algorithms we have used. Random Forest performs better than Naïve Bayes in this regard. In terms of accuracy, precision, recall, and f1 score Random Forest is the most superior model among the four, achieving the highest score for both smoker and drinker types Rand. By considering all that we can say Random Forest would be a great choice for predicting the types of smokers and drinkers.