<p>The hospitality sector generates a lot of data, which is added as written feedbacks and/or numerical ratings. Online travel agencies (OTAs) thrive on giving consumers a variety of options tailored according to their tastes and needs. In this research, we propose ranking of hotels according to customer provided reviews by incorporating machine learning and multi criterion decision making techniques. In this study, hotels located in Paris are shortlisted for the ranking purpose based on certain criterion. The reviews are initially pre-processed and a topic modelling technique Latent Dirichlet Allocation (LDA) has been utilized to extract important features. A total of seven features each having twenty keywords have been extracted through this technique. These features are then assigned weights using Maximum Entropy Minimum Variance Ordered Weighted Averaging (MEMV-OWA) method. For ranking, multi-criteria decision-modelling approach Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) has been utilized. Experts have been roped in to provide scores for the decision matrix of TOPSIS. To address the imprecision inherent in human judgement, picture fuzzy numbers have been used. The results of this procedure have ranked hotels located in Paris based on reviews provided by the customers. A comparison with two other MCDM techniques, VIKOR and PROMETHEE has been done to justify the proposed methodology. The findings of the research provide theoretical and practical implementations and are discussed in the paper.</p>