Sustainable Development Goals

Abstract/Objectives

This study proposes a framework that includes (1) the use of unsupervised natural language processing (NLP) methods to analyze user-provided data and (2) a recommendation system integrating deep learning applications to generate personalized solutions for customers. Thus, a customer is not only a service receiver but also an active data producer, helping form a value co-creation system with service providers. A case study involving traveling tourist recommendations validates the benefits of the proposed method. The main contribution of this research is the development of a personalized Smart PSS framework, which could achieve a win-win situation for all players in this framework.

Results/Contributions

This research developed a Smart PSS framework by incorporating deep learning models and human-provided data. The proposed system can handle unstructured data (i.e., text) in the context of a rapid-developing big data era. The framework consists of four steps, and the first is to discover hidden pain points of target customers via a modified customer journey map. Service providers are encouraged to put themselves in their customers’ shoes, rethink the entire original process, sketch out potential customers’ emotional journeys, and identify the pain points throughout this process. Following the framework, service providers then explore semantic data and suitable models to construct an improved service process, and a solution can be developed based on identified pain points, improving customer satisfaction. The contribution of this study includes the provision of a value co-creation Smart PSS framework between users and service providers, combining deep learning algorithms to deal with unstructured data, and a case study about questionnaire experiment to attest to the performance of the proposed framework.

Keywords

Smart Product Service SystemNatural Language ProcessingMachine LearningRecommendation System

Contact Information

邱銘傳
mcchiu@ie.nthu.edu.tw