Multi-Label Classification of Foreign Tourists' Opinions on Thailand Tourism Development

Simple item page
dc.contributor.authorSuramanka, Lalita
dc.contributor.authorHanskunatai, Anantaporn
dc.date.accessioned2026-08-06T10:47:13Z
dc.date.available2026-08-06T10:47:13Z
dc.date.issued2024-10-03
dc.description.abstractThe enhancement of tourism quality in Thailand through the understanding and utilization of foreign tourists' opinions presents challenges due to the extensive volume of data involved. This research proposes a two-fold approach to address this issue: (1) the development of an opinion classification model, and (2) the analysis of tourists' opinions through a dashboard. A dataset, compiled by the Tourism Authority of Thailand (TAT) and consisting of opinions from foreign tourists regarding areas for improvement in Thai tourism, was utilized. A total of 2,249 comments were collected. Experimental results demonstrate that the use of data augmentation, feature selection, Multi-label transformation using Classifier Chains, and the Random Forest classification model on the training dataset yields promising results with an accuracy rate of 80%, precision of 90%, recall of 81%, F1-score of 85%, and Hamming loss of 0.04. Analysis from the dashboard revealed the top three key areas for improvement: communication/language, traffic/public transportation, and cleanliness/hygiene.
dc.identifier.citationACM International Conference Proceeding Series, 32-38, 2024
dc.identifier.doi10.1145/3695220.3695227
dc.identifier.other2-s2.0-85207101861
dc.identifier.urihttps://dspace.kmitl.ac.th/handle/123456789/15938
dc.sourceACM International Conference Proceeding Series
dc.subjectFeature selection
dc.subjectMulti-label classification
dc.subjectOpinion mining
dc.subjectThai tourism
dc.titleMulti-Label Classification of Foreign Tourists' Opinions on Thailand Tourism Development
dc.typeConference Paper

Files

Collections

Research