<![CDATA[This study examines user reviews of the WhatsApp application as digital information objects that reflect user perceptions of digital information service quality. The rapid growth of communication platforms has generated large volumes of user-generated content, which requires systematic analysis and functions as a form of digital documentation. This research aims to evaluate how machine learning and deep learning approaches can support information evaluation through sentiment analysis of user reviews. A publicly available dataset of WhatsApp user reviews obtained from Kaggle was used as the data source. The research methodology consisted of text preprocessing, feature representation, sentiment classification, and performance evaluation. Support Vector Machine (SVM) was employed as a baseline machine learning method, while Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) models represented deep learning approaches. The experimental results show that deep learning models outperform the traditional approach, with CNN achieving the best performance across accuracy, precision, recall, and F1-score metrics. These findings indicate that deep learning-based sentiment analysis is effective in transforming large-scale user reviews into actionable information for evaluating digital information services. This study contributes to documentation and information science by demonstrating the role of artificial intelligence in analyzing user-generated digital documentation to support evidence-based decision-making in digital service development.]]>
