Title :
HUGE: an integrated system for human understandable granule extraction
Author :
Castellano, G. ; Fanelli, A.M. ; Mencar, C.
Author_Institution :
Dept. of Informatics, Bari Univ, Italy
Abstract :
An integrated system for the extraction of interpretable information granules from data is presented. The system, called HUGE (human understandable granule extraction), is designed as a highly reusable software framework that can embody several techniques for information granulation within a homogeneous user interface. HUGE includes a set of tools that enable both qualitative and quantitative evaluation of the results of a granulation technique. Visualization tools allow the user to assess graphically the properties of extracted information granules. Evaluation tools give the user the possibility to estimate numerically the quality of the derived granules. The usefulness of HUGE in supporting the user in a granulation process is shown through a case study concerning the design of an interpretable fuzzy inference system for predicting automobile fuel consumption.
Keywords :
automotive engineering; fuzzy systems; inference mechanisms; software reusability; software tools; user interfaces; automobile fuel consumption; fuzzy inference system; human understandable granule extraction; information granule extraction; integrated system; reusable software; user interface; visualization tool; Aggregates; Automobiles; Computer architecture; Data mining; Fuzzy systems; Humans; Software architecture; Software reusability; User interfaces; Visualization;
Conference_Titel :
Fuzzy Information Processing Society, 2005. NAFIPS 2005. Annual Meeting of the North American
Print_ISBN :
0-7803-9187-X
DOI :
10.1109/NAFIPS.2005.1548625