Open Conference Systems, ICQQMEAS2015

Font Size: 
Fitting Binomial Distribution to Online Ratings of Greek Tourism Attractions
Ioannis Dimotikalis, Stelios Papadakis

Last modified: 2015-09-24

Abstract


In this paper, we identify the distribution of the visitor’s online ratings for 100 tourism attractions in Greece. Provided that the visitors’ rating follows the Binomial distribution, we identify the parameter of Binomial distribution by using Particle Swarm Optimization (PSO). The objective of modeling is to investigate the possible bias on user ratings and the performance of fitting method compared to traditional Least Squares and Maximum Likelihood estimation. A novel statistical error measure, namely, “False Rate” is used for the rating data. We used a real world dataset of 100 Greek Tourism Attractions, downloaded from TripAdvisor website. The implications of the tourist attractions, the fitting performance and the five stars rating system is also being discussed

Full Text: PDF