Using Orbital Measurements to Determine Sex with Machine Learning
Author Information
Author(s): George Triantafyllou, George G. Botis, Maria Piagkou, Konstantinos Papanastasiou, George Tsakotos, Ioannis Paschopoulos, George K. Matsopoulos, Stavroula Papadodima
Primary Institution: National and Kapodistrian University of Athens
Hypothesis
Can orbital measurements from dried skulls be used to accurately estimate sex using machine learning?
Conclusion
Orbital measurements can serve as reliable predictors for automated sex estimation in forensic science.
Supporting Evidence
- The model achieved an overall test accuracy of 0.68.
- The precision of the model was 0.65, indicating a moderate level of false positives.
- The recall was 0.70, meaning 70% of positive cases were correctly identified.
- The F1 score was 0.675, suggesting a balanced performance between precision and recall.
- The area under the curve (ROC AUC) score was 0.72, indicating good class distinction.
Takeaway
Scientists measured parts of skulls to see if they could tell if a person was male or female using computers. They found that these measurements can help, but there's still room for improvement.
Methodology
Orbital measurements were taken from 92 dried skulls, and a Random Forest classifier was used to analyze the data.
Potential Biases
The model may be biased towards the majority class due to the unequal distribution of sexes in the sample.
Limitations
The sample size was small and imbalanced, which may affect the generalizability of the findings.
Participant Demographics
35 males and 57 females from Athens, Greece.
Statistical Information
P-Value
<0.001
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website