Pengembangan Chatbot Moderasi Bahasa Kasar Pada Platform Whatsapp Sebagai Solusi Otomatis Pengawasan Etika Berkomunikasi

The increasing intensity of communication on the WhatsApp platform often triggers the spread of toxic language, which is difficult for group administrators to monitor manually due to encryption constraints and time limitations. This study designs an automated moderation chatbot system by integrating n8n as the workflow manager and Google Gemini as the context analysis engine. Utilizing the System Development Life Cycle (SDLC) method with the Waterfall model, the system adopts a "Care and Control" approach by issuing tiered warnings before enforcing automatic removal (auto-kick) sanctions. Implementation results demonstrate that Google Gemini possesses superior stability and response time efficiency, averaging 4.39 seconds, which significantly outperforms other comparative models in terms of message processing speed and accuracy. Functionality testing using the Black Box method showed a 100% success rate across all primary features, including toxic word detection, the warning system, and command features. Furthermore, model performance testing based on Accuracy Metrics (Confusion Matrix) yielded absolute values of 100% for Accuracy, Precision, Recall, and F1-Score on the test data. The results of this study conclude that the implementation of a Gemini AI-based moderation chatbot is a highly effective, precise, and safe solution against false positives in maintaining communication ethics within WhatsApp groups.