Semantic analysis of news of self-inflicted deaths in Yucatán

Yucatán is one of the states in Mexico with high suicide rates for several years compared to the rest of the country. Because of this, local newspapers have emphasized publishing-related news. Many of them contain associated expressions that help characterize the text and define its semantics. This work presents the implementation of a text classification algorithm based on the semantic analysis of suicide news in Yucatán. Different classification models were built and evaluated, and the best results were obtained with Logistic Regression using TF-IDF. The algorithm was tested with 122 news items, of which it managed to identify 116 news items correctly, with effectiveness of 95.08%.