<![CDATA[ Background: Hair loss is a common problem that can affect a person’s self-confidence. Early prediction of the risk is important to help with more appropriate treatment.Objective: This study aims to apply the Na¨ıve Bayes algorithm to predict hair loss based on personal data and clinical factors such as age, gender, stress levels, hormones, and family history.Methods: The Na¨ıve Bayes method was chosen because it efficiently handles categorical data. The data used in this study were obtained from a public dataset available on the Kaggle platform, which contains individual information about the risk of hair loss.Result: The developed prediction model can classify risks based on various causal factors, but its performance is still low with an accuracy of 55.5%, AUC 0.593, and MCC 0.113.Conclusion: These results indicate that the model is unreliable for practical applications. The implication is that this system can be the basis for further development with more complex algorithms, the addition of clinical features, and stronger validation so that it can be applied effectively in medical contexts and personal consultations. ]]>
Copyrights © 2025
