Title :
A New Tool for Merging the Information Based on Clustering Methods
Author :
Garfias, Ing José Miguel Torres ; Molina, Antonio Orantes ; Flores, Jesús Linares ; Avalos, M. C Jorge L Barahona
Author_Institution :
Univ. Tecnol. de la Mixteca, Oaxaca, Mexico
Abstract :
This paper shows the results of a generic tool developed for unsupervised classification based on the methodology LAMDA (Learning Algorithm for Multivariable Data Analysis). The new tool is called GUILSI (Generic tool of classification Using the algorithm Lamda for Supervision of functional states and assignment of elements). It describes how the tool GUILSI was used to segment an image and to make the diagnosis of complex processes, such in a chemical plant. It also makes a comparison of the new tool made in this research with the existing tool: SALSA (Situation Assessement using Lamda classification Algorithm). The new tool consists of two modes: learning by historical data (offline) and classification of new data (online).
Keywords :
data analysis; learning (artificial intelligence); pattern classification; pattern clustering; GUILSI; LAMDA methodology; SALSA; chemical plant; clustering method; data classification; generic tool; historical data learning; image segmentation; lamda classification algorithm; lamda for supervision algorithm; learning algorithm for multivariable data analysis; situation assessement; unsupervised classification; Classification algorithms; Data models; Equations; Image segmentation; Mathematical model; Support vector machine classification; Vectors; Unsupervised classification; complex process; learning algorithm;
Conference_Titel :
Electronics, Robotics and Automotive Mechanics Conference (CERMA), 2011 IEEE
Conference_Location :
Cuernavaca, Morelos
Print_ISBN :
978-1-4577-1879-3
DOI :
10.1109/CERMA.2011.86
