IMPLEMENTASI SISTEM NOTIFIKASI MULTIPLATFORM DALAM MONITORING TINGKAT STRES MAHASISWA BERBASIS SKOR DASS-21

College students are highly susceptible to mental health issues due to high academic and social pressure, requiring an early detection system that can monitor stress levels quickly and responsively. This research aims to build and implement a multiplatform notification system capable of automatically sending early warnings through WhatsApp, Email, push notification, and Smartwatch, based on the evaluation results of students' DASS-21 scores using the Forward Chaining inference method. The system was designed using AppSheet as the input medium for the DASS-21 questionnaire, with the data then processed through the n8n platform as the automation hub that runs the Forward Chaining inference engine to classify depression, anxiety, and stress scores into five categories, supported by an AI Agent (Google Gemini) to generate contextual recommendations for students. The classification results are then automatically distributed to the four notification channels. System testing was conducted using Black-Box Testing with 8 scenarios, along with latency testing on WiFi and mobile data networks. The results show that the system was successfully built and all processes ran as expected, the Forward Chaining engine accurately classified the scores, and notifications to students functioned properly across all platforms, with delivery times well below the established SLA of 60 seconds. It was also found that notification delivery speed was more affected by real network conditions than by the type of connection used. Meanwhile, escalation notifications to counselors for severe and extremely severe stress categories have only been implemented at the rule-base level and have not been tested end-to-end, making this a direction for future development. This research is expected to provide a meaningful contribution to supporting proactive and sustainable early detection and monitoring of student mental health.