Abstract:

Social media has become a dominant platform for sharing opinions, and millions of users review and rate tourist destinations on tourism websites daily. Sentiment analysis of these reviews can provide valuable insights into the popularity of tourist spots, helping travelers make informed decisions on where to visit. This paper implements sentiment analysis using machine learning techniques. The dataset is sourced from multiple tourism review websites. A comparative study of feature extraction algorithms, namely CountVectorization and TFIDFVectorization, is conducted alongside classification algorithms such as Naive Bayes (NB), Support Vector Machine (SVM), and Random Forest (RF). The performance of these algorithms is evaluated based on metrics such as accuracy, recall, precision, and F1-score. The results show that TFIDFVectorization outperforms CountVectorization in improving the accuracy of classification algorithms for the given review dataset. Specifically, the combination of TFIDFVectorization and Random Forest achieves the highest accuracy of 86% on the research dataset used for sentiment classification of tourist place reviews.

Description:

.