<![CDATA[The Indonesian government is striving to enhance digital public services, including the Digital Identity Application (IKD) launched in 2022 by the Directorate General of Population and Civil Registration. Since its launch, IKD has received various responses from the public. User reviews on Google Play Store indicate a decline in ratings from June to December 2023. Review analysis is essential to understand user satisfaction, identify issues, and guide application improvements. This study aims to perform sentiment analysis on IKD user reviews using Bidirectional Long Short-Term Memory (Bi-LSTM) and Word2Vec methods. Bi-LSTM and Word2Vec are used to develop sentiment analysis from previous research that still used Machine Learning methods. This research is expected to contribute to the development of sentiment analysis models using Deep Learning for the IKD application. Review data was collected from the Google Play Store using scraping techniques for the period January-December 2023 and categorized into positive and negative. The Bi-LSTM model was trained with Word2Vec CBOW and Skip-Gram variations with dimensions of 100, 200, and 300. The results show that the combination of Bi-LSTM and Word2Vec CBOW with a dimension of 200 and a data split ratio of 80/20 produced the highest accuracy of 96.06%, with a precision of 96.44%, recall of 95.64%, and an f1-score of 96.04%. All combinations of Bi-LSTM and Word2Vec outperformed other Machine Learning algorithms.]]>
