A Machine Learning Approach for Predicting Nicotine Dependence

Abstract

An examination of the ability of machine learning methodologies in classifying women Waterpipe (WP) smoker's level of nicotine dependence is proposed in this work. In this study, we developed a classifier that predicts the level of nicotine dependence for WP tobacco female smokers using a set of novel features relevant to smokers including age, residency, and educational level. The evaluation results show that our approach achieves a recall of 82% when applied on a dataset of female WP smokers in Jordan.

Applicability of machine learning algorithm to predict the therapeutic intervention success in Brazilian smokers

“…Algorithms performance evaluation. In order to reduce outlier effects in the final decision of each algorithm, we calculated the mean values and standard-deviation (m±SD) of the following metrics: accuracy (ACCU), sensitivity (SENS), specificity (SPE), predictive positive value (PPV) and area under the receiver operating characteristic curve (AUC) [14,[39][40][41].…”

“…Previous studies show that for smokers, ML application showed good results in the empirical study of smoking cessation intervention in Korea (precision between 67.3% and 87.7%) [ 15 ], and nicotine dependence evaluation in Jordanian women (precision of 82.0%) [ 13 ]. However, in Brazilian healthcare services, the ML application for screening of patients is scarce and as far as we know it was used only in the cardiovascular risk evaluation [ 18 ], highlighting lack of screening of variables related to therapeutic intervention success (TIS) in Brazilian smokers using ML algorithms.…”

“…Machine learning (ML) can be a key resource to improve the therapeutic success of smokers, since it gets together the variables with specific characteristics, such as the treatment effects, level of nicotine dependence, and so on, which individually can contribute to the defined outcome. The combination of them gives us a diagnostic or prognostic of health problems [13][14][15], and it can be used to better comprehend the dataset structure associated with smoking quit and to construct user-friendly tools, such as risk equators [16][17][18][19].…”

Respiratory failure in COVID-19 patients a comparative study of smokers to nonsmokers

“…The selection process is based on the predictor features that are known [4], [5]. Feature selection (FS), also known as attribute selection, is an essential phase of building any predictive model [6], [7]. This is essential since the number of features could be large and others less informative.…”

“…Explaining research chronological, including research design, research procedure (in the form of algorithms, Pseudocode or other), how to test and data acquisition [7]- [17]. The description of the course of research should be supported references, so the explanation can be accepted scientifically [2], [6]. Figures 1-2 and Table 1 are presented center, as shown below and cited in the manuscript [7], [20]- [30].…”