Banca de QUALIFICAÇÃO: ARTHUR COSTA GORGÔNIO

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : ARTHUR COSTA GORGÔNIO
DATE: 22/10/2020
TIME: 15:00
LOCAL: Remoto
TITLE:
A Semi-supervised Method for Data Stream Applications

KEY WORDS:
Semi-supervised learning, classification in data stream, concept drift

PAGES: 73
BIG AREA: Ciências Exatas e da Terra
AREA: Ciência da Computação
SUBÁREA: Metodologia e Técnicas da Computação
SPECIALTY: Sistemas de Informação
SUMMARY:
Applications in the data streams domain receive a large volume of data quickly and there is a need to process them 
sequentially. A feature of these applications is that data can change during the process of using the model, in addition 
the number of instances whose label is known may not be enough to generate an effective model. In order to 
suppress the difficulty of the small number of instances labeled, learning can be used semi-supervised and, the use 
of classifier committees can assist in the detection of change of context. Thus, in this work, a method is proposed to 
perform the classification semi-supervised in tasks with continuous data flows, using a approach based on classifier 
committees. To assess the effectiveness of the proposal, empirical tests are carried out with five databases, a 
classifier committee semi-supervised and two different batch sizes, using accuracy as a metric, precision, recall and 
F-Score The preliminary results indicate that this proposal achieved better results compared to supervised 
approaches when uses batches with a larger number of instances.

BANKING MEMBERS:
Presidente - 1350250 - ANNE MAGALY DE PAULA CANUTO
Externo ao Programa - 4351681 - JOAO CARLOS XAVIER JUNIOR
Externa ao Programa - 2720574 - KARLIANE MEDEIROS OVIDIO VALE
Externo à Instituição - ARAKEN DE MEDEIROS SANTOS - UFERSA
Notícia cadastrada em: 08/10/2020 17:30
SIGAA | Superintendência de Informática - | | Copyright © 2006-2020 - UFRN - sigaa09-producao.info.ufrn.br.sigaa09-producao