Título: Uso de Técnicas de Mineração de Dados na Prevenção de Acidente Vascular Cerebral
Autores: Mello Junior, Harold;Figueiredo, Karla;Pereira, Antonio;Gabriela, Luiza
The present work compares the results of data mining processes on data from questionnaires answered by the public during campaigns to raise awareness about stroke held in 2015 and 2016. Stroke is one of the leading causes of death worldwide and the use of data mining techniques can help uncover associated risk factors and help prevent the occurrence of more cases. Four traditional classification algorithms were used in the database, which contained information of 592 individuals on the following parameters: socioeconomic, anthropometric, medical history, and knowledge of risks associated with stroke. The results show that classification improves when the participants’ knowledge of stroke, including risks, physiopathology, signs and symptoms, etc. are included in the database. The random forest and C4.5 algorithms provided the best classification outcomes about stroke risk with perfect 100% scores, followed by neural network and part with 95% and 97.66%, respectively.
Cerebral Vascular Accident (CVA);Risk Factors;Health Education;Data Mining;Classification Algorithms.
Código DOI: 10.21528/CBIC2017-129
Artigo em pdf: cbic-paper-129.pdf
Arquivo BibTeX: cbic-paper-129.bib