<![CDATA[The development of Artificial Intelligence (AI) has brought significant changes in the field of information and communication. The Ask-AI application is popular and has many reviews on the Google Play Store platform. The purpose of this study is to analyze user review sentiments towards the Ask-AI application and compare the performance of two text classification algorithms, namely Naive Bayes and Support Vector Machine (SVM) in classifying reviews into positive and negative sentiment categories. A total of 628 reviews were used as a dataset consisting of 314 positive reviews and 314 negative reviews. The dataset has gone through a text preprocessing stage including letter transformation (transform cases), tokenize, common word removal (stopword removal), and dictionary-based stemming. Data analysis using RapidMiner software and for model performance evaluation using the k-fold cross-validation approach which can provide more stable and representative results for the entire data. The evaluation results produce a performance value of the SVM algorithm which has very good performance. SVM produces an accuracy of 94.08%, a precision of 96.23%, a recall of 92.31%, and an Area Under Curve (AUC) value of 0.981. Meanwhile, the Naive Bayes algorithm provides an accuracy of 78%, a precision of 85.23%, a recall of 68.37%, and an AUC of 0.801. The results of the study indicate that the SVM method is superior to Naïve Bayes in classifying the sentiment of Ask-AI application user reviews because it can provide more accurate, consistent, and more sensitive classification results to variations in text data. It is hoped that this study can be a reference for choosing the optimal sentiment classification algorithm for AI-based application user review data.]]>
Copyrights © 2025
