The rapid expansion of the sharing economy has significantly transformed the hospitality and tourism industry, with peer-to-peer (P2P) accommodation platforms such as Airbnb redefining customer experiences and service expectations. This thesis examines the factors influencing customer retention in Vietnam’s P2P accommodation sector, with a particular focus on revisit intentions. Using a dataset of over 86,000 customer reviews collected from Airbnb listings across Vietnam, this study applies advanced Natural Language Processing (NLP) techniques, including Latent Dirichlet Allocation (LDA) for topic modeling and sentiment analysis with VADER and SentiWordNet. Customer feedback was labeled into four primary themes: room quality, service quality, hospitality, and surroundings, aligning with the theoretical framework of perceived value dimensions. By leveraging machine learning models such as Logistic Regression, Random Forest, and CatBoost, this study predicts customer retention behaviors based on thematic and sentiment-based insights. The findings highlight that room quality and hospitality are the most significant drivers of repeat bookings, supported by positive reviews and host responsiveness. Service quality and surroundings also contribute substantially, particularly in creating memorable experiences and ensuring customer satisfaction. Furthermore, the study demonstrates the effectiveness of sentiment-based labeling as an alternative approach to understanding retention behaviors. Integrating text feedback analysis with predictive modeling provides a robust framework for identifying actionable insights. This study contributes to the literature supporting customer retention in the sharing economy by addressing gaps specific to emerging markets such as Vietnam and provides practical recommendations for P2P accommodation providers to enhance guest experience, improve retention rates, and achieve sustainable growth in an increasingly competitive market.
Bản đồ thống kê
Thống kê nội dung
The rapid expansion of the sharing economy has significantly transformed the hospitality and tourism industry, with peer-to-peer (P2P) accommodation platforms such as Airbnb redefining customer experiences and service expectations. This thesis examines the factors influencing customer retention in Vietnam’s P2P accommodation sector, with a particular focus on revisit intentions. Using a dataset of over 86,000 customer reviews collected from Airbnb listings across Vietnam, this study applies advanced Natural Language Processing (NLP) techniques, including Latent Dirichlet Allocation (LDA) for topic modeling and sentiment analysis with VADER and SentiWordNet. Customer feedback was labeled into four primary themes: room quality, service quality, hospitality, and surroundings, aligning with the theoretical framework of perceived value dimensions. By leveraging machine learning models such as Logistic Regression, Random Forest, and CatBoost, this study predicts customer retention behaviors based on thematic and sentiment-based insights. The findings highlight that room quality and hospitality are the most significant drivers of repeat bookings, supported by positive reviews and host responsiveness. Service quality and surroundings also contribute substantially, particularly in creating memorable experiences and ensuring customer satisfaction. Furthermore, the study demonstrates the effectiveness of sentiment-based labeling as an alternative approach to understanding retention behaviors. Integrating text feedback analysis with predictive modeling provides a robust framework for identifying actionable insights. This study contributes to the literature supporting customer retention in the sharing economy by addressing gaps specific to emerging markets such as Vietnam and provides practical recommendations for P2P accommodation providers to enhance guest experience, improve retention rates, and achieve sustainable growth in an increasingly competitive market.
