Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Jurnal Sistem Komputer dan Informatika (JSON)

Implementasi Support Vector Machine untuk Kendali Lampu Ruangan Adam, Chairul; Hidayati, Rahmi; Suhardi, Suhardi
Jurnal Sistem Komputer dan Informatika (JSON) Vol 5, No 3 (2024): Maret 2024
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/json.v5i3.6964

Abstract

The use of lights in daily activities is a necessity for everyone. Lights are useful as lighting when the room has minimal light, so lights are needed in the room in the house. If there are many rooms in a house, there are also many lights used. Often users forget to turn on and off the lights in each room if done manually. An automatic control is needed to make it easier for users to turn on and off lights according to user habits. Support Vector Machine is used in this in this study to classify the condition of lights based on user habits when turning on and off the lights. User habit data used for 15 days with a total of 620 data. Testing was carried out for 3 days with a total of 72 data. The results of system testing obtained the accuracy value of kitchen lights by 87.50%, accuracy of main room lights by 95.83%, accuracy of second room lights by 91.67%, accuracy of living room lights by 95.83%, accuracy of terrace lights by 95.83%, and accuracy of toilet lights by 94.44%.
Sistem Deteksi Dini dan Pemadaman Kebakaran Otomatis di Rumah Berbasis IoT Menggunakan NodeMCU ESP32 Saputra, Dimas Bagus; Hidayati, Rahmi; Suhardi, Suhardi
Jurnal Sistem Komputer dan Informatika (JSON) Vol 5, No 3 (2024): Maret 2024
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/json.v5i3.6974

Abstract

Fire is a serious disaster that causes significant losses in terms of both material assets and loss of life, especially in densely populated residential areas. Fires that occur in one house can easily spread to other houses. To prevent this issue, an early fire detection and automatic fire suppression system has been developed for homes based on the Internet of Things (IoT) using NodeMCU ESP32. This system has the ability to detect early signs of fire, gas, and temperature conditions, and takes automatic action when potential fire hazards are detected. Additionally, the system sends real-time push notifications to users' Android applications to provide warning messages. Test results indicate that the fire sensor can detect flames up to a distance of 130 cm with an average notification delay of approximately 2,37 seconds. The MQ-2 gas sensor can detect butane gas up to a distance of 170 cm with an average notification delay of about 2,84 seconds, while the DHT22 temperature sensor has an average accuracy of approximately 98,52% with a notification delay of about 2,03 seconds.