Claim Missing Document
Check
Articles

Found 1 Documents
Search

Penerapan Algoritma Deep Learning Pada Chatbot Untuk Pelayanan Informasi Website Institut Teknologi Batam Menggunakan Metode Long Short Term Memory Budi Prasetio; Luki Hernando; Joni Eka Candra; Muhammad Jufri
JURNAL QUANCOM: QUANTUM COMPUTER JURNAL Vol. 4 No. 1 (2026): Juni 2026
Publisher : LPPM-ITEBA

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62375/j5fawr38

Abstract

In the current digital era, information plays an important role in various sectors, including information services in higher education. To handle repetitive questions that are often asked by p        rospective new students to academics or campus marketing, chatbots that utilize Artificial Intelligence (AI) technology and natural language processing are an efficient solution. This research aims to implement deep learning algorithms, especially the Long Short-Term Memory (LSTM) method, on chatbots to improve information services on the Batam Institute of Technology (ITEBA) website. The chatbot was developed using the Python programming language and the Flask framework, creating a well-integrated web interface and supported by the necessary deep learning libraries. The evaluation results show that the LSTM model is able to consistently reduce loss values ​​and increase accuracy, with loss values ​​reaching 63% and accuracy reaching 85%. This reflects excellent performance and generalization ability. Further testing revealed that the chatbot had significant capabilities in understanding and responding to various user questions with high accuracy and relevant answers. This chatbot not only increases efficiency and accuracy in information services, but also provides a better experience for prospective students and other users, and significantly optimizes the information service process at the Batam Institute of Technology.