Sentiment analysis and artificial neural networks-based econometric models for tourism demand forecasting (CROSBI ID 667831)
Prilog sa skupa u zborniku | izvorni znanstveni rad | međunarodna recenzija
Podaci o odgovornosti
Folgieri, Raffaella ; Baldigara, Tea ; Mamula, Maja
engleski
Sentiment analysis and artificial neural networks-based econometric models for tourism demand forecasting
Purpose – This is the second step of a previous paper (Folgieri et al, 2017), where we modelled and applied a backpropagation Artificial Neural Network (ANN) to forecast tourists arrivals in Croatia. Tourism is a very important sector of current Countries’ economies, and forcasting assumes even more an significant issue to lead the local tourist offer. In this context, early prediction on the tourist inflow represents a challenge as it is an opportunity in developing tourist income. Applying a Machine Learning Method for Decision Support and Pattern Discovery such as ANN, represents an occasion to achieve a greater accuracy if compared to results usually obtained by other methods, such as Linear Regression. Design – In this paper, we extended the model of the previously used backpropagation Artificial Neural Network, including data from sentiment analysis collected through social networks on the Internet. Methodology –The accuracy of the neural network has been measured by the Mean Squared Error (MSE) and compared to results obtained applying the ANN without data coming from the sentiment analysis. Approach – Our approach consists in combining ideas from Tourism Economics and Information Technology, in particular Artificial Intelligence methods, such as Machine Learning and sentiment analysis, throught the Artificial Neural Networks (ANN) we used in our study. Findings – The results showed that including also data from sentiment analysis, the neural network model to predict tourists arrivals outperforms the previous obtained results. Originality of the research –The idea to use ANN as a Decision Making tool to improve tourist services in a proactive way or in case of unexpected events is innovative. Adding data from sentiment analysis, we can add also tourists' preferences so considering collective intelligence and collective trends as factors which could influence a prediction.
Artificial Neural Networks ; Econometrics ; Forecasting ; Artificial Intelligence ; Machine Learning ; Prediction
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
Podaci o prilogu
88-97.
2018.
objavljeno
Podaci o matičnoj publikaciji
Tourism and Hospitality Industry
Milohnić, Ines ; Smolčić Jurdana, Dora
Opatija: Faculty of Tourism and Hospitality Managemenet
2623-7407
Podaci o skupu
24. bijenalni međunarodni kongres Turizam i hotelska industrija: trendovi i izazovi = 24th Biennial International Congress Tourism & Hospitality Industry - Trends and Challenges
predavanje
26.04.2018-27.04.2018
Opatija, Hrvatska
Povezanost rada
Ekonomija