Knowledge representation using hypergraph in product development
Product designers in the information age have access to an increasing amount of raw data that they can use to inform their design process. However, the knowledge in product development is complex and massive involving heterogenous entities and relations that require contextual interpretations. At th...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
Nanyang Technological University
2021
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/150492 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | Product designers in the information age have access to an increasing amount of raw data that they can use to inform their design process. However, the knowledge in product development is complex and massive involving heterogenous entities and relations that require contextual interpretations. At the same time, the product development cycle has become a multi-stage, iterative and multidisciplinary collaborative process. Thus, an effective knowledge management system is required to structure the vast amount of data and allow for efficient communication throughout the design process. Knowledge graphs have been increasingly used to support knowledge representation especially for knowledge in linguistic form because of its wide applicability and emphasis on relations between entities. However, conventional knowledge graphs are limited in representing multi-ary relations commonly found in practical datasets. To address this limitation, a hypergraph-based method for knowledge management in the domain of product development was proposed. Hypergraphs are a generalization of conventional graphs with the key difference being the use of hyperedges which is much more effective in representing multi-ary relations. In this report, a case study of user generated knowledge from a 3D printer forum was conducted to demonstrate the validity of the proposed hypergraph-based knowledge representation method. |
---|