Classificação de Risco de Suicídio Utilizando Análise deLinguagem Natural
Abstract
In recent decades, the high rates of suicide worldwide have been drawing the attention of government agencies and society in general. Strategies for treatment and prevention have been formulated, however, there is a great difficulty in detecting people in situations of risk. In this context, the present work proposes a method for suicide risk classifier using Natural Language Processing (NLP), which seeks to identify suicidal intentions in text messages. The classifier is a Naive Bayes learning algorithm that delivers 70.45% accuracy, 54.2% acceptance rate for suicidal messages and 95% for non-suicidal.
References
Botega, Neury José; WERLANG, Blanca Susana Guevara; CAIS, Carlos Filinto da Silva; MACEDO, Mônica Medeiros Kother. Prevenção do comportamento suicida. Psico, v. 37, n. 3, p. 5, 2006.
Hettige, Nuwan C. et. al. Classification of suicide attempters in schizophrenia using sociocultural and clinical features: A machine learning approach. General Hospital Psychiatry, v. 47, p. 20-28, 2017.
Lewis, David D. Naive (Bayes) at forty: The independence assumption in information retrieval. In: European conference on machine learning. Springer, Berlin, Heidelberg. p. 4-15, 1998.
Mccallum, Andrew; Nigam, Kamal. A comparison of event models for naive bayes text classification. In: AAAI-98 workshop on learning for text categorization. p. 41-48, 1998.
Ministério da Saúde. Portaria Nº 1.876, de 14 de agosto de 2006. Institui Diretrizes Nacionais para Prevenção do Suicídio, a ser implantadas em todas as unidades federadas, respeitadas as competências das três esferas de gestão.
Organização Mundial da Saúde. Departamento de Saúde Mental e de Abuso de Substância. Prevenção do Suicídio um Recurso para Conselheiros. Genebra, 2006. Disponível em: [http://www.who.int/mental_health/media/counsellors_portuguese.pdf]. Acesso em: 01 jan. 2018.
Organização Mundial da Saúde. Mental health: Suicide data. 2016. Disponível em: http://www.who.int/mental_health/prevention/suicide/suicideprevent/en/. Acesso em: 01 jan. 2018.
Pak, Alexander; Bernhard, Delphine; Paroubek, Patrickand; Grouin, Cyril. A combined Approach to emotion Detection in suicide notes. Biomedical informatics insights, v. 5, n. Suppl 1, p. 105-114, 2012.
Phuc, Do; Phung, Nguyen Thi Kim. Using Naïve Bayes model and natural language processing for classifying messages on online forum. In: Research, Innovation and Vision for the Future, 2007 IEEE International Conference on. IEEE, 2007. p. 247-252.
Spasić, Irena; Burnap, Pete; Greenwood, Mark; Arribas-Ayllon, Michael. A naïve bayes approach to classifying topics in suicide notes. Biomedical informatics insights, v. 5, n. Suppl 1, p. 87-97, 2012.
