A study of Tobacco use and mortality by data mining

Abstract

The use of data mining to address the issue of people who consume tobacco and other harmful substances for their health has led to a significant dependence among smokers, which over time causes illnesses that may result in the addict's death. As a result, the research's goal is to apply a data mining study whose findings showed that the confidence intervals are less than 0.355. However, the lift and conviction in the last three rules are also lower, making it unlikely that these rules will be followed. On the other hand, the knowledge discovery in data bases method was used. It consists of the following stages: data selection, preparation, data mining, and evaluation and interpretation of the results. To that end, comparisons of agile data mining methodologies like crisp-dm, knowledge discovery in data, and Semma are also done. As a result, using specific criteria, dimensions are segmented to allow for the differentiation of these methodologies. As a result, a comparison graph of models such as naive Bayes, decision trees, and rule induction is used. To sum up, it can be said that the rules of association apply to men, the number of admissions, and the cancers that can be brought on by smoking. Also, the percentage of male patients admitted with cancers that can be brought on by smoking Last but not least, the number of admissions and cancers that can be brought on by smoking