Classificação probabilística baseada em análise de padrões

• Autor(a) principal: Merschmann, Luiz Henrique de Campos
• Data de Publicação: 2007
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da Universidade Federal Fluminense (RIUFF)
• Texto Completo: https://app.uff.br/riuff/handle/1/17300
• topic: Computação, Biologia computacional, Mineração de dados, Classifificação, Bioinformática, Otimização combinatória; Computational biology, Data mining, Classifification, Bioinformatics, Combinatorial optimization; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO

Classification is a data mining tast that has been useful in several application areas, particularly, in bioinformatics. The genomic revolution has resulted in an explosive growth of biological data generated by the scientific community. With the aim of storing all of these biological information, biological databases were created. The need for computational tools for analysing biological data becomes evident, resulting in the application of data mining methods in this field. The work developed in this thesis is related to classification task and, initially, to its application to bioinformatics. The initial goal is to present a computationally efficient method for protein classification capable of yielding highly accurate results, outperforming the results obtained by previous works. The good results in terms of accuracy and time performance obtained by the proposed method show its potential for the protein classification problem. In addition, aiming to construct a suitable classifier for several kinds of applications, the method proposed for the protein classification problem was extended, becoming appropriate and efficient for several databases associated with different applications.