Data Mining Applied to Analysis of Contraceptive Methods Among College Students

Area of Science:

• Public Health
• Data Mining
• Epidemiology

Background:

• Understanding contraceptive method utilization is crucial for reproductive health.
• University populations often present unique sexual health challenges and vulnerabilities.
• Data mining offers powerful tools for analyzing complex health behavior data.

Purpose of the Study:

• To apply Data Mining techniques to analyze contraceptive method usage patterns.
• To identify specific vulnerabilities in contraceptive method use within a university setting.
• To profile contraceptive use in a southern Brazilian university population.

Main Methods:

• Utilized a comprehensive database on sexuality from a university population.
• Employed Data Mining to generate analytical rules regarding contraceptive use.
• Validated study findings using accuracy, sensitivity, specificity, and ROC curve analysis.

Main Results:

• Generated rules largely align with existing global literature and epidemiological data.
• Identified significant points of vulnerability in contraceptive method use among university students.
• Achieved high or comparable validation measures (accuracy, sensitivity, specificity, ROC AUC) to similar studies.

Conclusions:

• Data Mining effectively profiles contraceptive use and reveals critical vulnerabilities in university populations.
• Findings underscore the need for targeted reproductive health interventions in higher education.
• The study's methodology and results demonstrate robust scientific validity.