investigating the evolution of hotel experience from unstructured data: A topic modeling application on travelers' online reviews.

In this age of information technology, collective information is increasingly being digitized and shared through many platforms such as blogs, social media, news and online review systems. Given the unparalleled growth of user generated content (UGC) in the form of text, both researchers and users have the chance to extract valuable information from them. In the context of the lodging industry, while guests opinions represent a double-edged sword affecting the future performance of targeted hotels, they can also be deployed by lodging operators to enhance customer satisfaction. If hotels want to survive and endure the changing business environment, they have to adapt and exploit the proliferation of new information technology systems, such as online reviews platform. In fact, even though the functionality of these systems seems to be beyond firms control, an opportunity for hotel managers to take advantage of the platform still exists. This commitment requires a strong investment in computerized managerial support decision making, as well as machine learning techniques for the extrapolation of hidden patterns in text data. Among them, text mining offers an unprecedented advantage. With the aim of providing hotel managers a way to translate unstructured text data into useful knowledge, this study explores which hotel dimensions primarily affect hotel guests experience through an application of topic modeling algorithms. Moreover, it focuses on how online review platforms and data-drive techniques can help hotel managers realize whether re-thinking and transforming their core competences into dynamic capabilities foster customer experience, and consequently hotel performance.