Application of the Fuzzy System for an Emotional Pattern Generator

Round 1

Reviewer 1 Report

This paper proposed a Mamdani fuzzy system based method to help to customize the car interior trim elements by customers' emotions.

Generally, the paper is well-written, and the research topic is eye-catching. However, it is not clearly demonstrated how to evaluate the proposed method.

It will be better to design an experiment to evaluate the proposed method quantitatively instead of a simple description. If possible, please add it to the paper.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper is generally well-written and the authors are aware of many trends and previous works.

The authors understand "good design" mostly as "aesthetic appearance", but I know many designers who would not agree and who would argue that a deep understanding of the users' culture and value system,
amongst others, is an essential ingredient too. And quality of interaction is essential too, for example. Some modesty would be good here regarding the claims.

An important branch of research which does not appear in the reference list is
the work on Design & Emotion by Pieter Desmet and Paul Hekkert.

I applaud the use of generators, I think that too many designers just work by WYSIWYG editors and do not know the power of compilers (generators).

The usage of Fuzzy logic makes me think of the somewhat similar fuzzy specialized Mondrian-art generator described in Feijs,L. (2019). A program for Victory Boogie Woogie. Journal of Mathematics and the Arts, 13(3), 261-285.

The weak point of the paper is the presentation of the validation study.
The experiment of section 5 appears interesting, but the authors
do not give enough details to check whether the claims are true or not.
We need at least a stepwise description of the procedure:
first the test subject sees this, then we ask him that,etc.
How many tasks were assigned to each subject (only the Elegant-Classic-Masculine, or others too)? Then we need the results: plots, averages and standard deviations of the correct and wrong outcomes.
Ideally, we even have a statistical analysis (eg using SPSS) to tell whether the outcome is statistically significant, or just a trend.

The repairwork on the validation is mandatory.
If that is done, I recommend to accept (otherwise not).

Author Response

Please see the attachment.

Author Response File: Author Response.pdf