Currently, the process of evaluating suicide is highly subjective, which limits the efficacy and accuracy of prevention efforts. Artificial intelligence (AI) has emerged as a mean of investigating large datasets to identify patterns within ‘big data’ that can determine the factors on suicide outcomes. Here, we used Spark NLP for Healthcare to extract the topic from Spanish press and social media texts. However, news media articles lack of suicide tags.

Using tweets with hashtags related to suicide, we trained a neuronal model that identifies if a given text has a suicide-related topic. Our results suggest a high level of impact of suicide cases in the media, and an intrinsic thematic relationship of suicide news. These results pave the way to build more interpretable suicide data from the media, which may help to better track, understand its origin, and improve prevention strategies. The dataset, model, and code are all free and open-source.