<![CDATA[Digital transformation requires customer service to be fast, responsive, and continuously accessible. To address this demand, this study presents the development of an AI-based chatbot employing the Long Short-Term Memory (LSTM) algorithm to enhance customer support for PLN. LSTM was chosen due to its effectiveness in capturing conversational context and understanding natural language patterns. The development process includes data preprocessing, model training, and performance evaluation using metrics such as accuracy, precision, recall, and F1-score. Experimental results on 133 test samples demonstrate an accuracy of 82.71%, with an average precision of 82%, recall of 77%, and F1-score of 77%, indicating reliable model performance. The chatbot is designed to handle common customer inquiries, including billing information, service disruptions, and other general services. This innovation is expected to improve PLN’s operational efficiency while delivering faster, more personalized, and dependable customer service, aligning with the demands of the digital era.]]>
Copyrights © 2025
