Pengembangan Aplikasi Chat Bot Whatsapp, Telegram, dan Website Laporan untuk Instansi Pemerintah Disperkim Pool Seroja Kota Semarang Berbasis Natural Language Processing

Abstract

The rapid development of information and communication technology has encouraged government agencies to improve the quality of public services through digital transformation. This research focuses on developing a multi-platform chatbot application integrated with WhatsApp, Telegram, and a web-based reporting system to support the public service operations of DISPERKIM POOL Seroja, Semarang City. The main issues identified include limited access to information, dependency on office hours, and increased workload for staff in handling repetitive inquiries from the public. Using a methodology that includes needs analysis, system design, implementation, and testing, the chatbot was developed using Natural Language Processing (NLP) to understand user intent in natural Indonesian language and provide accurate, real-time automated responses. The implementation results show that the system performs effectively, achieving an intent recognition accuracy rate of over 85% and an average response time of less than three seconds. User testing also indicates high satisfaction levels, as the chatbot significantly simplifies access to information and facilitates faster reporting processes. Overall, the system enhances public service efficiency, reduces the burden on DISPERKIM staff, and supports the government's digital transformation agenda toward smart and responsive public services.