PENGEMBANGAN APLIKASI CHATBOT WHATSAPP UNTUK PELAYANAN AKADEMIK DI TEKNIK ELEKTRO UMM DENGAN PENDEKATAN NLP

In the higher education environment, the need for fast and easily accessible academic information services is increasingly important. However, most academic information delivery is still conducted conventionally and lacks interactivity. This study aims to develop an artificial intelligence (AI)-based chatbot using a Natural Language Processing (NLP) approach through two different platforms: Eva.id and Gradio. The first chatbot was developed using Eva.id and integrated with the WhatsApp application to provide automated responses to student inquiries. Meanwhile, the second chatbot was built using Google Colab and Gradio, utilizing the multi-qa-MiniLM-L6-cos-v1 NLP model to process documents and match questions based on meaning. This system was applied to the Electrical Engineering Department at Universitas Muhammadiyah Malang. Based on a survey conducted with 24 respondents, the Eva.id-based chatbot achieved an average score of 3.3 out of 4, with a user satisfaction level of 78.3%. These results indicate that document-based and NLP-powered chatbots can serve as an effective alternative solution to support academic information services in higher education institutions.