Introduction The information that the architects deal with is in more often qualitative rater than quantitative. This already indicates some difficulty that may emerge while trying to process the information from this domain. To obtain numerical data is essential for processing and therefore, for exact sciences like pure engineering sciences it occurs quite natural to deal with numerical data, while in soft sciences, like architecture, this may not seem so natural due to the qualitative aspects of its data. Expressions such as: bright color, light room, large space are some of these examples. To deal with such information the emerging technologies of the last decade can provide an important aid. Such as, for example, the soft computing (Jang, et al 1997; Kaynak, et al 1998; Chen et al 1999; Azvine, et al 2000;). The tools for data analysis in soft computing are mainly referred to as neuro-fuzzy systems involving fuzzy logic and neural networks. With these tools it is feasible to discover the most important aspects that determine the quality of a space and, thereafter, to define exactly the shape of a function of one fuzzy concept to the other. The data used in this paper is obtained from an extensive inquiry based on a questionnaire regarding four different underground stations in the Netherlands, which is processed by soft computing techniques. After information processing, knowledge is modeled where the relationship between various design aspects is provided. In total there are 43 design aspects that are considered, which are directly related to comfort and safety of these particular stations. It is important to mention that for underground stations the aspects of comfort and safety are those that eventually determine the quality of a space. The following section provides a quick scan of aspects that were dealt with in the questionnaire. Thereafter, a brief introduction of soft computing techniques applied in this research for knowledge modeling is provided. Finally, the knowledge elicitation from the knowledge is done Quantifying the Qualitative Design Aspects DURMISEVIC, Sanja; CIFTCIOGLU, Özer; SARIYILDIZ, Sevil Delft University of Technology, Faculty of Architecture, The Netherlands http://www.bk.tudelft.nl/informatica/toi/www/
[1]
D. Broomhead,et al.
Radial Basis Functions, Multi-Variable Functional Interpolation and Adaptive Networks
,
1988
.
[2]
Detlef Nauck,et al.
Intelligent Systems and Soft Computing: Prospects, Tools and Applications
,
2000
.
[3]
Hjp Harry Timmermans,et al.
Modeling Consumer Perception of Public Space in Shopping Centers
,
1999
.
[4]
Özer Ciftcioglu,et al.
Knowledge management by information mining
,
2001
.
[5]
E. Mizutani,et al.
Neuro-Fuzzy and Soft Computing-A Computational Approach to Learning and Machine Intelligence [Book Review]
,
1997,
IEEE Transactions on Automatic Control.
[6]
Lotfi A. Zadeh,et al.
Outline of a New Approach to the Analysis of Complex Systems and Decision Processes
,
1973,
IEEE Trans. Syst. Man Cybern..
[7]
Romedi Passini,et al.
Wayfinding in Architecture
,
1984
.
[8]
Jack L. Nasar,et al.
Fear of Crime in Relation to Three Exterior Site Features
,
1992
.
[9]
Okyay Kaynak,et al.
Computational Intelligence: Soft Computing and Fuzzy-Neuro Integration with Applications
,
1998,
NATO ASI Series.
[10]
Chuen-Tsai Sun,et al.
Functional equivalence between radial basis function networks and fuzzy inference systems
,
1993,
IEEE Trans. Neural Networks.