<![CDATA[The rapid advancement of technology has brought significant changes in various aspects of life, including the education sector. As an educational institution, Nurul Jadid University must adopt the latest technology to enhance efficiency and service, particularly in responding to the increasing volume of inquiries and information needs from the public and parents before enrolling their children. A chatbot, as part of Natural Language Processing (NLP) based on Artificial Intelligence (AI), is designed to interact with users through text or voice, providing fast, accurate, and continuous service. The Long Short-Term Memory (LSTM) algorithm in deep learning is utilized for text data prediction and classification. In this research, the data consists of tags, patterns, and responses obtained manually from the official Nurul Jadid University website and then preprocessed to develop the chatbot model. The core component of this model is the embedding layer, which assigns vector values to each word in the processed text data. The model training results indicate an accuracy of 99.32% and a loss of 12.57%, demonstrating that the model performs well without overfitting or underfitting, making it suitable for testing and deployment. Thus, the LSTM-based chatbot serves as an effective virtual assistant to help the public, prospective students, and current students access information more easily and efficiently.]]>
Copyrights © 2025
