Data mining approach investigates Western-style restaurant hospitality management in Taiwan

Abstract

Purpose In terms of service hospitality, recent discussions of value-in-use from the perspective of service-dominant logic have focused on the customer’s determination of value and control of the value creation process. The purpose of this paper is to extend these discussions by exploring the value creation process in the Western-style restaurant in Taiwan, which is developed value-in-eat creation for restaurants. In Taiwan, Western-style restaurants are as popular as Chinese restaurants because of globalization and cultural integration. However, to local restaurateurs and managers, managing a Western-style restaurant in terms of localization and hospitality on value-in-eat creation presents both academic and practical issues. Thus, this paper aims to investigate Western-style restaurant hospitality management alternatives on the value-in-eat creation process in Taiwan using a data mining approach. Design/methodology/approach Based on a market survey, a total of 1,187 questionnaires was incorporated into a database. The questionnaire design is divided into 7 parts with 35 items. All questions are designed as nominal and ordinal (not the Likert scale) scales. Data mining approach, including cluster analysis and association rules, cluster analysis is investigated possible customer profiles and association rules is implemented to explore customer preference patterns and rules on the value-in-eat creation process. Findings Data mining results show two patterns including Pattern 1: meal patterns and customer preferences for restaurant hospitality management and Pattern 2: customer relationship management (CRM) for restaurant hospitality management that customer profiles and preferences on meal patterns, service patterns and CRM are engaged to suggest effective Western restaurant hospitality management alternatives, such as proper bundles for restaurant types, meals, exotic atmosphere and services of hospitalities in terms of a value-in-eat creation process. Originality/value To the best of the authors’ knowledge, this study is the first study to investigate consumers’ behaviors in Western-style restaurants using the measurement of nominal and ordinal scale for questionnaire development and further to implement a data mining approach on selected data samples. In addition, this study illustrates the patterns/rules of Taiwan customer preferences that best explain the knowledge of how to manage Western-style restaurants from the perspective of customer hospitality using data mining.

Market segmentation based on customer experience dimensions extracted from online reviews using data mining

“…Choi et al (2020) collected survey data from 421 customers on six predefined attributes (marketing, food quality, service quality, dietary concerns, reputation and overall experience) to group food truck customers into four categories based on their priorities – health-conscious, cost-oriented, taste-oriented or convenience-oriented. Liao et al (2021) used k -means cluster analysis with survey data to find three clusters: peace of mind and diversity group (<25-year-old, concerned with waiting time for meals, seating comfort, premeal service), nostalgia and simple group (26–35-year-old, concerned with long wait times and retro atmosphere), the trend of foundation segment (<25-year-old, concerned with restaurant tidiness and meal duration). Most of these segmentation studies have used a survey-based approach that may also have limitations due to factors such as the size of the data set, scope of inquiry (type of restaurants, locations) and bias resulting from self-reported responses.…”

“…Second, existing literature has predominantly focused on a predetermined set of criteria to assess customer preferences which may limit a comprehensive understanding of customer needs (Bujisic et al , 2014; Cheng et al , 2010). Third, numerous studies have used either survey-based or transaction-based approaches to segment consumers (Liao et al , 2021; Ponnam et al , 2011; Tan and Lo, 2008), but these methods can be limited by the availability of data and resource requirements (Moon et al , 2021). Finally, for products like food that cannot be inspected before consumption, consumers typically rely on sensory information such as taste and freshness to make their purchase decisions (Wang et al , 2023).…”