Emotional Analysis of Online Evaluation of Tourism Based on Improved Bayesian Algorithm
Miao Dong, Weichang Jiang, Rongrong Sun · 2024
Aiming at tourists' emotional analysis can better activity tourists' real feedback, most of the tourists' comment data are generated by tourists' initiative after the tour, and its authenticity is higher. Analyzing the differences in tourists' perceptions and influencing factors in tourism can help scenic spots develop differentiated service strategies and improve tourist satisfaction. Taking four 5A-level scenic spots as research objects, machine learning algorithms and content analysis are utilized to classify the emotional of tourism reviews. Firstly, Python crawler is used to obtain the data of tourism reviews, and the acquired data are cleaned and processed; secondly, a plain Bayesian classifier is improved and constructed according to the feature attributes of the samples, and then the data of the test set are classified based on the Dalian University of Technology emotional Dictionary. Conclusion It is proved that the algorithm can effectively improve the accuracy of text emotional classification, reveal the perception differences of tourists from the level of perceived content and emotional features, and effectively avoid the effect that the attribute whose conditional probability is equal to 0 has its a posteriori probability of the whole class equal to 0.