Notícias

Banca de QUALIFICAÇÃO: ORRANA LHAYNHER VELOSO DE SOUSA

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
DISCENTE: ORRANA LHAYNHER VELOSO DE SOUSA
DATA: 10/06/2022
HORA: 14:00
LOCAL: google meet
TÍTULO: Ensemble of Classifiers for Multilabel Clinical Text Categorization in Portuguese
PALAVRAS-CHAVES: Machine Learning Clinical data, ensemble, embeddings, classification
PÁGINAS: 10
GRANDE ÁREA: Ciências Exatas e da Terra
ÁREA: Ciência da Computação
RESUMO:

The widespread adoption of medical document management and recording systems has generated a large volume of unstructured data containing abbreviations, ambiguous terms, and typing errors. These factors make manual categorization an expensive, time-consuming, and error-prone task. Thus, the automatic classification of medical data into informative clinical categories can substantially reduce the cost of this task. In this context, this work aims to evaluate the use of an ensemble of classifiers of clinical texts to differentiate them into prescriptions, clinical notes, and exam requests. For this, we used three combinations of embeddings in the representation of the text. Then, we used the Support Vector Machine, the Random Forest, and the Multilayer Perceptron algorithms to create the classification ensemble. After that, we predict the final ensemble label through a voting approach. The results achieved with this methodology were promising, reaching an accuracy of 1.00, kappa of 0.99, and F1-score of 1.00. As a result, our approach allows for automatic and accurate classification of the content of clinical texts, achieving better categorization results than the unique approaches evaluated.



MEMBROS DA BANCA:
Interno - 2025063 - ROMUERE RODRIGUES VELOSO E SILVA
Externo ao Programa - 1794678 - VICTOR EULALIO SOUSA CAMPELO
Externo à Instituição - RAFAEL TORRES ANCHIÊTA - IFPI
Notícia cadastrada em: 17/05/2022 09:25
SIGAA | Superintendência de Tecnologia da Informação - STI/UFPI - (86) 3215-1124 | © UFRN | sigjb05.ufpi.br.instancia1 28/03/2024 14:27