Algoritmos para Mineração de Dados Utilizando Regressão Logistica - Aplicação em Bioinformatica Estrutural
Several problems in bioinformatics are rooted in the search for relationships between components that are not detectable at first glance, similar to the retrieval of latent in- formation in search engines such as Yahoo, Google, etc. Artificial intelligence techniques used by these machines transform...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis doctoral |
| Estado: | Versión publicada |
| Fecha de publicación: | 2023 |
| País: | Brasil |
| Institución: | Universidade Federal de Minas Gerais (UFMG) |
| Repositorio: | Repositório Institucional da UFMG |
| Idioma: | portugués |
| OAI Identifier: | oai:repositorio.ufmg.br:1843/64777 |
| Acceso en línea: | http://hdl.handle.net/1843/64777 |
| Access Level: | acceso abierto |
| Palabra clave: | bioinformática regressão Logística mineração de dados Bioinformática Modelos Logísticos Mineração de Dados |
| Sumario: | Several problems in bioinformatics are rooted in the search for relationships between components that are not detectable at first glance, similar to the retrieval of latent in- formation in search engines such as Yahoo, Google, etc. Artificial intelligence techniques used by these machines transform enormous amounts of data into discovery and kno- wledge. Among the resources available in the area, Logistics Regression occupies a little explored place, but has a lot to offer. In this work, we show how traditional Logistic Regression, with some of the modifications suggested here, can be used successfully in a class of bioinformatics problems; namely, that involving the three-dimensional structure of proteins. This application joins others, where the versatility of the technique could be assessed. It has already been explored in classifiers built from microarrays, in the reuse of drugs, in virtual screening of ligands, in the search for druggable targets and in phylogenetic trees. |
|---|