共 28 条
ESTIMATING PERCEPTIONS OF CONSUMERS USING AI: A CASE STUDY FOR GERMAN AND UK TRAVELERS
被引:0
|作者:
Kejo, Starosta
[1
]
Budz, Sonia
[1
]
Krutwig, Michael
[1
]
机构:
[1] Bucharest Univ Econ Studies, Bucharest, Romania
来源:
BASIQ INTERNATIONAL CONFERENCE: NEW TRENDS IN SUSTAINABLE BUSINESS AND CONSUMPTION 2018
|
2018年
关键词:
Tourism;
online media;
sentiment analysis;
machine learning;
TOURISM DEMAND;
SENTIMENT;
D O I:
暂无
中图分类号:
F [经济];
学科分类号:
02 ;
摘要:
This empirical study analyzes the different perceptions of popular European tourist destinations in German and UK media reports. There are huge differences in the behavior of travelers from different countries, their preferred tourist destinations, accommodations, etc. Most of these differences can be explained by the differences in disposable income, exchange rates, gravity models that simply account for the travel distance, and some more. While all these metrics are reasonable explanatory variables, this study tries to elaborate whether there are further differences that originate from the general perception of a country. To measure the differences in the perceptions, we analyzed the differences in the sentiments of the online media in Germany and UK toward potential travelling destinations and compared this information with tourist arrivals in these countries. The sentiments in the media were measured by artificial neural network software that analyzes the mood of the mass media. The results revealed that there were indeed different perceptions in the German and UK online media for some European tourist destinations and these differences were reflected in tourist arrivals. These results suggested that there were either news on different topics or that the media in Germany and UK valued the same topics in different ways. This information is useful for businesses in the tourism and hospitality sectors as well as for country promoters and news provider businesses to investigate the source and the impact of the information gap to optimize their operations.
引用
收藏
页码:181 / 188
页数:8
相关论文