Development of an Interface for Computing and
Analyzing Multidimensional Perceptual Space
Lin-Lin Chen and Jeng-Weei LinDepartment of Industrial and Commercial Design
National Taiwan University of Science and Technology
1
Introduction
Product perceptual space [ Green 1989; Moore 1993] is often used to visualize
consumers' perception of the various products in the market, to determine
consumers' satisfaction with the existing products, and to seek opportunities for
new products, or directions to improve existing products. Product perceptual
space can be constructed by using a number of methods, including multidimensional scaling (or MDS) [ Kruskal 1978; Schiffman 1981].When using MDS for similarity analysis or preference analysis, the designers
often face the following difficulties [ Lin 1995]:| 1. | Collecting similarity or preference data from surveys, and transforming the collected data into the input formats required by the various MDS programs are tedious and error-prone. |
| 2. | In the perceptual space generated by MDS programs, stimulus objects are represented symbolically, as numbers or names, rather than graphically, as icons or images. For design related applications, such a representation does not provide sufficient information for interpreting the perceptual space. To visualize the distribution of stimulus objects in the perceptual space, designers often create a graphical representation from the output of MDS programs by hand, or by using a separate drawing program. |
| 3. | To interpret the meaning of each axis in the perceptual space, designers need to visualize ideal vectors that represent subject preferences, as well as the projections of points onto the ideal vectors. Again, designers usually need to draw these vectors and to find projections of points based on the output of MDS programs. |
-621-
