Configuring products with natural language: a simple yet effective approach based on text embeddings and multilayer perceptron

ABSTRACT

Product configurators are recognised as critical toolkits enabling customers to co-create products with companies. Most available product configurators require customers to select suitable product attributes from predefined options. However, customers usually find the selection processes frustrating due to their lack of product knowledge. In view of the fact that customers often express their needs in imprecise and vague natural language, we define a new needs-based configuration mechanism and propose an implementation approach based on text embeddings and multilayer perceptron. Specifically, we leverage the massive amount of product reviews by encoding them into text embeddings. A multilayer perceptron is trained to map text embeddings to product attribute options. Experiment results indicate that the mapping has good generalisation capability to map customer needs into product configurations. The performance of our approach is comparable to that of deep learning-based approaches but with much higher efficiency in terms of computational complexity. Our needs-based configuration thus provides a quick and effective means of facilitating product customisation. It also demonstrates an innovative way of utilising customer resources in unstructured text to co-create products with companies.

KEYWORDS:

• Product customisation
• needs-based configuration
• text embedding
• multilayer perceptron
• natural language processing

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 The complexity of MLP is O(LK), where L is the number of words in the text corpus and K is the dimension of word embeddings. The complexity of the concatenation approach is O(2LK)= O(LK) as O(f(n))= O(f(cn)), where c is a constant. Thus, the element-wise average, max pooling and concatenation have the same computational complexity.

Additional information

Funding

This work was supported by Hong Kong Research Grant Council Faculty Development Scheme [grant number UGC/FDS14/E06/18].

Notes on contributors

Yue Wang

Yue Wang received the B.S. degree in Electronics and M.E. in Information Science from Peking University, Beijing China, and the Ph.D. degree in Industrial Engineering from The Hong Kong University of Science and Technology. He is an Associate Professor in the Department of Supply Chain and Information Management, the Hang Seng University of Hong Kong. His research interests include machine learning, natural language processing, design and manufacturing informatics, healthcare informatics, and supply chain management. His Erdos number is 3.

Xiang Li

Li Xiang received the B.S. degree in Internet of Things Engineering from Central South University, Chang Sha, China in 2018 and M.Sc. degree in Information Technology from The Hong Kong University of Science and Technology, Hong Kong in 2019. He is a Research Assistant in Department of Supply Chain and Information Management, The Hang Seng University of Hong Kong. His research interests focus on the data mining and natural language processing.

Linda L. Zhang

Dr. Zhang is currently a Professor of Operations Management in Department of Operations Management at IÉSEG School of Management (LEM-CNRS 9221), Lille-Paris, France. She obtained her BEng and Ph.D. degrees in Industrial Engineering in 1998 and 2007, respectively. Her research interests include mass customisation, product and production configuration, sustainable supply chain management, healthcare operations management, etc. On these areas, she has published a number of articles in international refereed journals, such as Decision Support Systems, IIE Transactions, IEEE Transactions on Engineering Management, European Journal of Operations Research, International Journal of Production Economics, etc.

Daniel Mo

Daniel Y. Mo received the Ph.D.degree in industrial engineering and logistics management from the Hong Kong University of Science and Technology (HKUST), Hong Kong, in 2013. He is currently an Associate Professor with the Department of Supply Chain and Information Management, The Hang Seng University of Hong Kong. He has authored or co-authored research papers in international journals such as the IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING, IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, etc. His current research interests include service parts operations management, logistics systems optimisation, design of intelligence systems, and Lean Six Sigma implementation. Dr.Mo was a recipient of the Kayamori Best Automation Paper Award Finalist for one of his research works. He is a professional of Certified Six Sigma Black Belt by the Hong Kong Society for Quality and a Member of the Institute of Industrial Engineers.