Please use this identifier to cite or link to this item:
Title: An ontology learning system for customer needs representation in product development
Authors: Chen, Xingyu
Chen, Chun-Hsien
Leong, Kah Fai
Jiang, Xing
Issue Date: 2012
Source: Chen, X., Chen, C.-H., Leong, K. F., & Jiang, X. (2012). An ontology learning system for customer needs representation in product development. The International Journal of Advanced Manufacturing Technology, 67(1-4), 441-453.
Series/Report no.: The international journal of advanced manufacturing technology
Abstract: The intense competition and high failure rate of product introduction necessitate a deeper understanding of customer needs for product design. The conventional process of interpreting customer statements relies on imprecise information, making it highly unlikely, if not impossible, to acquire accurate need statements for the front-end process of product development. To deal with this problem, an ontology-learning customer needs representation (OCNR) system is proposed in this paper. The system uses natural language processing tools to preprocess customer statements. The customer needs ontology is then established based on the key concepts and their relations that are extracted from the customer statements. A set of need statements are then generated using the established customer needs ontology. A word property-based method is proposed to extract more nontaxonomic relations. A case study was conducted to illustrate the proposed approach. Results of this study suggest that the customer needs ontology derived from the proposed OCNR system contains more semantics than those obtained from the existing ontology learning systems, and, therefore, might be able to generate more accurate need statements.
DOI: 10.1007/s00170-012-4496-2
Fulltext Permission: none
Fulltext Availability: No Fulltext
Appears in Collections:MAE Journal Articles

Citations 20

Updated on Jul 16, 2020

Web of ScienceTM
Citations 20

Updated on Mar 4, 2021

Page view(s)

Updated on Sep 25, 2022

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.