Scholarlink

Drug abuse is a rapidly growing public health issue, particularly among today’s youth. As a result, the problem of determining an individual’s risk of drug consumption and abuse is critical and new. This research focuses on detecting drug users based on demographic data while keeping personality characteristics constant. The dataset was retrieved from the University of California, Irvine (UCI) repository, which contains standard benchmark data. The dataset consists of 12 features that can be classified in 18 ways. Although the dataset can be broken down into 18 categories, this study focuses on three: amphetamine users, LSD users, and alcohol users. This research helps to determine the best machine learning method for the binary classification challenge. The statistics show that there are more than 90% of responses in all categories. Experimental results demonstrate that the Gradient Boosting classifier performs well in all three characteristics with a diverse range of features.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Demographic and Personality-Based Drug Abuse Prediction Using Machine Learning Models

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • R. Srinivasan,
  • Sharmila Devi Chandariah,
  • S. Karthick,
  • P. Anitha Rajakumari

摘要

Drug abuse is a rapidly growing public health issue, particularly among today’s youth. As a result, the problem of determining an individual’s risk of drug consumption and abuse is critical and new. This research focuses on detecting drug users based on demographic data while keeping personality characteristics constant. The dataset was retrieved from the University of California, Irvine (UCI) repository, which contains standard benchmark data. The dataset consists of 12 features that can be classified in 18 ways. Although the dataset can be broken down into 18 categories, this study focuses on three: amphetamine users, LSD users, and alcohol users. This research helps to determine the best machine learning method for the binary classification challenge. The statistics show that there are more than 90% of responses in all categories. Experimental results demonstrate that the Gradient Boosting classifier performs well in all three characteristics with a diverse range of features.