Fundamentals of speech recognition
Fundamentals of speech recognition
Pattern recognition and classification in time series analysis
Applied Mathematics and Computation
A new time series classification approach
Signal Processing
The process of knowledge discovery in databases
Advances in knowledge discovery and data mining
Multidimensional curve classification using passing—through regions
Pattern Recognition Letters - Special issue on pattern recognition in practice VI
Selecting Models from Data: AI and Statistics IV
Selecting Models from Data: AI and Statistics IV
Time Series Analysis: Forecasting and Control
Time Series Analysis: Forecasting and Control
| Hi-index | 0.00 |
A new hybrid methodology for Knowledge Discovery in Serial Measurement (KDSM) and the results of applying it to psychiatry are presented in this paper. In the application domain where serial measurements are repeated and very short (i.e. very few parameters), traditional measuremethods for series analysis are inappropriate. Moreover, some information is non-serial but is closely connected to serial measurements. For this reason, common statistical analysis (time series analysis, multivariate data analysis ...) and artificial intelligence techniques (knowledge based methods, inductive learning) used independently provide often poor results because of the characteristics above and it is necessary a suitable way of analyzing these situations. KDSM is built as an hybrid methodology, specially designed to obtain knowledge from repeated very short serial measurement, in order to overcome the limitations of Artificial Intelligence or Statistics techniques. Novel knowledge about electroconvulsive therapy behavior was obtained once KDSM was applied to this specific domain. Thus, KDSM gives a possible solution to a knowledge problem.