KDSM Methodology for Knowledge Discovery from Ill-Structured Domains Presenting Very Short and Repeated Serial Measures with Blocking Factor

Authors:
Jorge Rodas;Karina Gibert;J. Emilio Rojo
Affiliations:
-;-;-
Venue:
CCIA '02 Proceedings of the 5th Catalonian Conference on AI: Topics in Artificial Intelligence
Year:
2002

Citing 3
Cited 0

Data mining

Data mining
Time Series Analysis: Forecasting and Control

Time Series Analysis: Forecasting and Control
Electroshock Effects Identification Using Classification Based on Rules

ISMDA '01 Proceedings of the Second International Symposium on Medical Data Analysis

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper is an introduction of Knowledge Discovery in Serial Measurement (KDSM) methodology for analyzing repeated and very short serial measures with a blocking factor in ill-structured domains (ISD). KDSM arises from the results obtained in a real application of psychiatry (presented in the previous issue of CCIA [11]). In this application domain, common statistical analysis (time series analysis, multivariate data analysis...) and artificial intelligence techniques (knowledge based methods, inductive learning), employed independently, are often inadequate due to the intrinsic characteristics of ISD. KDSM is based on both the combination of statistical methods and artificial intelligence techniques, including the use of clustering based on rules (introduced by Gibert in 1994).