Human-computer interaction and neural networks in architectural design - A tool for design exploration

Design research has demonstrated that neural networks are able to support creativity. However, there are two main problems with using neural networks in design. One is how you interact with such systems. The second relates to the integration between neural network techniques and other approaches. This paper will describe an integrated model in which those problems are addressed. The resulting system provides an interface in which the neural network output is translated into textual and graphic representations that can play a meaningful role in the design process.