Penerapan Chatbot Berbasis Natural Language Processing Untuk Layanan Informasi SPMB di SMPS Plus Fajar Sentosa

Abstrak

The New Student Admission System (SPMB) information service at SMPS Plus Fajar Sentosa still experiences response delays because it depends on committee members who also work as teachers, while the absence of standardized answers may lead to inconsistent information. This study aims to design and develop a Natural Language Processing (NLP)-based chatbot using the Naive Bayes method to provide fast, accurate, and consistent SPMB information through a website and WhatsApp. The system was developed using the Rapid Application Development (RAD) method, consisting of requirements planning, user design, construction, and cutover, while data were collected through observation, interviews, and literature study. User questions are processed through text preprocessing and classified using Naive Bayes to determine answers according to the appropriate information category. Functional testing showed that the system features operated according to the designed scenarios. User Acceptance Testing (UAT) involving 28 respondents, consisting of 26 users and 2 administrators, obtained overall scores of 87.24% from users and 95.00% from administrators, both of which were categorized as strongly agree. These results indicate that the chatbot can accelerate information services, improve answer consistency, and assist administrators in managing SPMB information. Therefore, the developed system is feasible for use as an SPMB information service medium at SMPS Plus Fajar Sentosa.

KeywordsChatbot, Natural Language Processing, Naive Bayes, SPMB, Rapid Application Development