A Machine Learning-Based Virtual Sensor System for Temperature Prediction in Forest Environments

Internet of Things (IoT) sensors are difficult to maintain because of the frequent replacement of components and batteries, which requires a lot of maintenance costs and manpower. To address these challenges, this study proposes a virtual sensor system that predicts temperature in forest environments by combining IoT sensor data with meteorological data. The system optimizes prediction performance by applying various regression models, ensuring high accuracy and reliability in real-time environmental monitoring. Future research will focus on developing models to predict additional environmental variables and explore the scalability of the virtual sensor system.