A multi-objective PSO approach of mining association rules for affective design based on online customer reviews

ABSTRACT

Affective design is an important aspect of new product development that can enhance customer satisfaction of new products. Previous studies generally conducted customer surveys based on questionnaires and interviews to collect customers’ views and preferences of affective design of products. However, the process could be time-consuming and the survey data does not contain much sentiment expression. Presently, a large number of online customer reviews on products can be found on various websites that contain rich information of customer opinions and expectations. However, the generation of useful information based on online customer reviews for affective design has not been addressed in previous studies. In this paper, a methodology for generating association rules for supporting affective design based on online customer reviews is proposed which mainly involves opinion mining of affective dimensions from online customer reviews and association rule mining based on multi-objective particle swarm optimisation (PSO). Opinion mining is adopted to analyze online reviews and conduct sentiment analysis for affective dimensions. Based on the mined information and morphological analysis of products, a multi-objective PSO approach is proposed to generate association rules that depict the relationships between affective dimensions and design attributes. A case study was conducted to illustrate the proposed methodology.

KEYWORDS:

• Affective design
• opinion mining
• association rule mining
• multi-objective PSO

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

K. M. Yu http://orcid.org/0000-0002-5187-9006

Additional information

Funding

The work was supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China [grant number POLYU517113] and also partially supported by a grant from The Hong Kong Polytechnic University [grant number G-YBMX].