This study investigates customer reviews to assess the quality of services offered by eco-friendly accommodation facilities, with the aim of supporting the advancement of sustainability practices in tourism development. Reviews were collected from major hospitality platforms using web scraping techniques. The data underwent pre-processing, including text normalization steps such as lowercasing, removal of stop words, punctuation, and numerical values. To convert the textual data into a numerical format suitable for analysis, Term Frequency–Inverse Document Frequency (TF-IDF) was applied. This technique assigns weights to words based on their contextual relevance, enhancing key terms while minimizing redundancy, thereby improving sentiment classification. A Random Forest machine learning model was employed to classify review sentiment and extract insights into customer perceptions. The results underscore the economic and regional value of sustainable tourism by highlighting the high appreciation of tourist customers for the quality of services in sustainable accommodations and demonstrate the potential of machine learning to reveal emerging trends and inform the best practices in eco-friendly hospitality management.