<![CDATA[The Legal Documentation and Information Network (JDIH) website of Tabanan Regency is a digital public service platform that provides access to regional regulations and legal documents. Evaluating website quality is important to ensure that the services provided are able to meet user needs and expectations. Previous studies on website evaluation generally applied either WebQual 4.0 or sentiment analysis separately, while studies integrating both approaches remain relatively limited. Therefore, this study combines WebQual 4.0 and machine learning–based sentiment analysis to provide a more comprehensive website evaluation. This study employed a mixed-methods approach involving 120 respondents and 120 user comments. Website quality evaluation was conducted using the WebQual 4.0 framework through three dimensions: usability, information quality, and service interaction. Sentiment analysis was performed using preprocessing techniques, TF-IDF feature extraction, and the Random Forest algorithm. The results showed that all WebQual dimensions achieved excellent categories, with usability obtaining the highest mean score of 4.487. Positive sentiment dominated user comments at 59.2%, while the Random Forest model achieved an accuracy rate of 79%. This integrated approach provided a more comprehensive evaluation and identified several technical aspects that still require improvement, including system performance, feature optimization, and interface consistency.]]>
