Development of a Supervised Algorithm for District Classification Based on Mental Health Prevalence in the Population

Abstract

The purpose of this scientific article was the creation of a supervised algorithm to categorize districts based on the frequency of mental health problems in their population. The research comprised six essential phases: understanding the problem, data preprocessing, feature selection, model training, model evaluation, and results analysis. The initial stage involved acquiring a deep understanding of the problem and its objectives. Data from mental health examinations were collected from the National Open Data Platform of the Peruvian Government. The proposed methodology has potential applicability in various contexts and countries, allowing a more precise understanding of the prevalence of mental health problems at the local level and facilitating decision-making in interventions and resource allocation. Validation revealed that the Support Vector Machine and Deep Learning models outperformed the logistic regression and decision tree models in terms of accuracy and overall performance. These findings underscore the potential of machine learning techniques to address mental health challenges at the local level. In summary, the results emphasize the importance of using data analytics and machine learning to address mental health issues and provide insights for future research.