Related Experiment Video
Updated: Sep 16, 2025

Assessment of Sexual Behavior of Male Mice
Published on: March 5, 2020
Machine Learning Identifies Sexual Behavior Subgroups Among Men Who Have Sex with Men in Switzerland
Luisa Salazar-Vizcaya1, Dunja Nicca2, Vanessa Christinet3
1Department of Infectious Diseases, Inselspital Bern University Hospital, University of Bern, Anna-Seiler-Haus, Geschoss J, 3010, Bern, Switzerland. luisapaola.salazarvizcaya@insel.ch.
Researchers identified distinct sexual behavior subgroups among men who have sex with men (MSM) using machine learning. First visit data effectively predicted subgroup membership, enabling tailored sexual health strategies.
Area of Science:
- Public Health
- Computational Epidemiology
- Behavioral Science
Background:
- Sexual behavior is complex and varies significantly across individuals.
- Understanding these diverse patterns is crucial for developing effective public health interventions and sexual health messaging.
- Men who have sex with men (MSM) represent a key population for targeted sexual health initiatives.
Purpose of the Study:
- To develop and validate a machine learning methodology for characterizing distinct sexual behavior subgroups within the MSM population.
- To assess the predictive power of initial visit data for identifying these subgroups.
- To inform the development of customized sexual health messages for specific subgroups.
Main Methods:
- Hierarchical clustering was used to group 2349 HIV-negative MSM based on longitudinal sexual behavior data.
- A random forest classification model was employed to predict subgroup membership using data from the first visit.
- Data were collected from Swiss sexual health counseling centers between November 2016 and April 2019.
Main Results:
- Six distinct subgroups with unique sexual behavior trends were identified, deviating significantly from overall population trends.
- Two subgroups (37% of participants) were associated with over 70% of increases in condomless anal intercourse with non-steady partners, group sex, and high numbers of partners.
- First visit data accurately predicted subgroup membership, with prediction accuracy ranging from 64% to 86%.
Conclusions:
- The study successfully identified specific sexual behavior subgroups among MSM, highlighting the heterogeneity within this population.
- Initial visit data can reliably predict subgroup membership, offering a practical approach for early identification.
- This methodology provides an algorithmic tool for creating targeted and personalized sexual health messages, improving public health outreach and sexual health outcomes.
Related Concept Videos
The Y Chromosome Determines Maleness
Evolution
Around 300 million years ago, the two sex chromosomes diverged from two identical autosomal chromosomes. Over time, the Y chromosome has lost most of its genes, shrinking in size....
Criticisms of the Evolutionary Perspective
Evolutionary psychology provides one explanation for these findings, suggesting...
Sex-linked Disorders
Sexually Transmitted Infections
Natural Selection and Mating Preferences
Females, due to their biological roles in conception, pregnancy, and nursing,...
Relationship Formation

