Title :
Knowledge discovery based on importance of features
Author :
Hiroshi, S. ; Kazunori, M.
Author_Institution :
Grad. Sch. of Eng., Kanagawa Inst. of Technol., Yokohama, Japan
Abstract :
This paper proposes a system which datamines time series classification knowledge leading by a discovery of feature patterns. In the case of classification, prediction accuracy is an important point, and to build a human understandable model is another essential issue. To satisfy these requests, our system runs in two stages. In the first stage, the system discovers important feature patterns which are useful for identifying data. For this purpose, we propose a feature importance measure which is called FI. The second stage builds a decision tree that determines class membership based on the feature patterns. We explain how these two stages are harmonized in the entire process.
Keywords :
data mining; pattern classification; time series; data identification; data mines time series classification; feature pattern discovery; knowledge discovery; prediction accuracy; Accuracy; Decision trees; Educational institutions; Feature extraction; Support vector machines; Time series analysis; Training data; automatic improvement; classification; feature discovery; knowledge extraction; time series data;
Conference_Titel :
Computers & Informatics (ISCI), 2012 IEEE Symposium on
Conference_Location :
Penang
Print_ISBN :
978-1-4673-1685-9
DOI :
10.1109/ISCI.2012.6222662