Login or signup to connect with paper authors and to register for specific Author Connect sessions (if available).

Decoding Trust: LLM in Provider Topic Modeling
Md Jabir Rahman, Samira Rahman, Md Juber Rahman
In this study, we explore how Airbnb hosts describe themselves online by harnessing the power of Latent Dirichlet Allocation (LDA) to uncover the unique aspects they highlight in their bios. First, utilizing an advanced deep learning framework (AB-BiLSTM) and a series of t-tests, we assessed the trustworthiness of hosts based on their bio content, revealing distinct variances across different host pairings. Next, the LDA analysis identified five predominant themes per host category that were filtered utilizing the power of a large language model, resulting in six unique topics across three host categories. Our findings reveal that the most successful hosts are highly customer-focused and exemplify the essence of hospitality and care. Conversely, less successful hosts tend to focus on the practical aspects of their rental service. Our study sheds light on the strategic self-presentation of Airbnb hosts and offers valuable insights for enhancing host effectiveness and guest satisfaction.

AuthorConnect Sessions

No sessions scheduled yet