An overview of data mining representation techniques

Current data mining systems and techniques are used with the objective of finding future values for datasets, leaving aside what could be even more useful information, by not exploring deeper into the outputs of such techniques. This paper presents an overview of the major representation and visuali...

ver descrição completa

Detalhes bibliográficos
Autor principal: Costa, Domingos (author)
Outros Autores: Portela, Filipe (author), Santos, Manuel (author)
Formato: conferencePaper
Idioma:eng
Publicado em: 2019
Assuntos:
Texto completo:http://hdl.handle.net/1822/67845
País:Portugal
Oai:oai:repositorium.sdum.uminho.pt:1822/67845
Descrição
Resumo:Current data mining systems and techniques are used with the objective of finding future values for datasets, leaving aside what could be even more useful information, by not exploring deeper into the outputs of such techniques. This paper presents an overview of the major representation and visualization techniques available and usable to bring forth all the potential associated with data mining techniques, especially those of a more complex nature, like artificial neural networks and support vector machines, with the objective of clarifying and raise awareness of the potential underlaying data mining techniques. The study uncovered a series of methods for this, such as a variety of techniques that help demonstrate artificial neural networks' visual model.