Reinforcing Tourism Post-pandemic Through a Natural Language Processing Data Analysis

Tourism is one of the sectors of the economy (global, regional, local) that was particularly heavily affected by Covid-19. As registered in many tourism departments from many countries, their financial incomes decreased since some of them were directly depended on the tourism sector. Sentiment analysis has emerged as a widely used method to understand the emotions and sentiments expressed by humans in text, to enhance the overall experience in a specific situation. The objective of this paper is to use advanced data analytics methods and natural language processing (NLP) techniques to extract comments from visitors about a tourist area, particularly those posted on Twitter during different periods of time. To identify what they considered as interesting sites in a tourist area, sentiment analysis is applied. The research focuses on tweets posted within a 3 km. radius from the center of Central Park in New York. This valuable information about the tourist’s opinions and interests can be used to gain insight information that can be used by the area management to take advantage of this to implement strategies that attract people's attention. The proposed methodology extracts tweets related to the study area and processes them by applying Natural Language Processing, then the texts are classified by a logistic regression algorithm. The relevance of the approach is the implementation of NLP and machine learning to include technology in the touristic area's management through the data analysis, and that the methodology could be replicated in different areas as long as the data exists.