Institutional Repository [SANDBOX]
Technical University of Crete
EN  |  EL

Search

Browse

My Space

Forecasting the success of a new tourism service by a neuro-fuzzy technique

Atsalakis Georgios, Atsalaki Ioanna, Zopounidis Konstantinos

Full record


URI: http://purl.tuc.gr/dl/dias/78FA6BDC-8096-41DB-9AC7-2D1F6133AE37
Year 2018
Type of Item Peer-Reviewed Journal Publication
License
Details
Bibliographic Citation G. S. Atsalakis, I. G. Atsalaki and C. Zopounidis, "Forecasting the success of a new tourism service by a neuro-fuzzy technique," Eur. J. Oper. Res., vol. 268, no. 2, pp. 716-727, Aug. 2018. doi: 10.1016/j.ejor.2018.01.044 https://doi.org/10.1016/j.ejor.2018.01.044
Appears in Collections

Summary

This paper presents a novel approach to forecasting the success of a newly launched service in tourism by using a hybrid intelligence system called the Adaptive Neuro Fuzzy Inference System (ANFIS). Recent studies have addressed the problem of modeling the success of a newly launched service by using different methods including artificial intelligence and model-based approaches. The ANFIS combines both the learning capabilities of a neural network and the reasoning capabilities of fuzzy logic to give enhanced forecasting capabilities, as compared to using a single methodology alone. Data collected through a questionnaire that concerns the variables of developing a new service in tourism have been used as inputs to the model. A new technique that is achieved by using a method that cycles through all the inputs and builds ANFIS models has been used for input reduction and input selection. The final model has been trained by leaving out a part of the data. The model was then evaluated by the data that were left out. The forecasting accuracy of the ANFIS model is evaluated by calculating well-known performance measures. The results have shown that ANFIS provides a prudent way to capture uncertainty in relationships among input variables and output variables to forecast the successful launch of a new tourism service. A comparative analysis with other methodologies confirms the superiority of the proposed approach.

Services

Statistics