|
Title:
|
EXTENSION OF DATA MINING ALGORITHM WITH FUZZY LOGIC FOR BI SYSTEM |
|
Author(s):
|
Angélica Urrutia, Christian Ortiz, Claudio Gutiérrez, José Galindo |
|
ISBN:
|
978-989-8533-14-2 |
|
Editors:
|
Hans Weghorn and Pedro Isaías |
|
Year:
|
2012 |
|
Edition:
|
Single |
|
Keywords:
|
Data mining, fuzzy data, fuzzy C 4.5 algorithm, ETL, BI Pentaho |
|
Type:
|
Short Paper |
|
First Page:
|
353 |
|
Last Page:
|
358 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
This work aims to incorporate linguistic labels in a layer of the classic architecture of Business Intelligence (BI) for Data Mining (DM), to make possible an extension in the analysis of the obtained results, considering the percentage of classified instances, confusion matrices and estimated errors. It starts with the evaluation of the exposed problematic, followed by a design of BI architecture for DM giving solution to the classification and prediction of quantitative management indicators. A data labeling technique in fuzzy logic is used through a trapezoidal function to give membership degrees to data. Once all of the problems and variables are solved, the solution is implemented through the process of knowledge extraction known as KDD, and a comparison of obtained results against classic logic is made. |
|
|
|
|
|
|
