How features embedded in eWOM predict hotel guest satisfaction: an application of artificial neural networks

ABSTRACT

Electronic-word-of-mouth (eWOM) plays a crucial role not only in analyzing customer decision-making but also in enhancing the revenue and brand image of service providers in the market. Understanding guest satisfaction is one of the important aspects because it augments the guest experience and their future intention to book. Many studies carried out in past applied regression-based and qualitative methods for analyzing the satisfaction levels of guests, but data analytics methods are still lacking in the literature and how these types of approaches can be applied to benefit hoteliers has not been completely explored. The objective of this study is to identify the important features that predict hotel guest satisfaction (HGS) through the ANN-based HGS prediction model. The feature importance with respect to guest satisfaction has also been estimated and discussed. We have discussed the managerial and theoretical implications of our research. Recommendations for the hospitality industry are also provided.

摘要

电子口碑（eWOM）不仅在分析客户决策方面起着至关重要的作用，而且在提高服务提供商在市场上的收入和品牌形象方面也起着至关重要的作用. 了解客人的满意度是一个重要的方面，因为它增加了客人的体验和他们未来的预订意向. 在过去的研究中，很多研究都是采用回归分析和定性分析的方法来分析客人的满意度，但是数据分析方法在文献中仍然缺乏，而且这些类型的方法如何应用于酒店经营者还没有完全的探索. 本研究的目的是通过基于人工神经网络的饭店顾客满意度预测模型，找出预测饭店顾客满意度的重要特征. 并对特征对顾客满意度的重要性进行了估计和讨论. 我们讨论了我们研究的管理和理论意义. 同时也为酒店业提供了建议.

KEYWORD:

• ANN
• eWOM
• text analytics
• guest satisfaction
• feature importance
• online reviews