The occurrence of wildfires often results in significant fatalities. As wildfires are notorious for their high speed of spread, the ability to identify wildfire at its early stage is essential in quickly obtaining control of the fire and in reducing property loss and preventing loss of life. This work presents a machine learning wildfire detecting data pipeline that can be deployed on embedded systems in remote locations. The proposed data pipeline consists of three main steps: audio preprocessing, feature engineering, and classification. Experiments show that the proposed data pipeline is capable of detecting wildfire effectively with high precision and is capable of detecting wildfire sound over the forest’s background soundscape. When being deployed on a Raspberry Pi 4, the proposed data pipeline takes 66 milliseconds to process a 1 s sound clip. To the knowledge of the author, this is the first edge-computing implementation of an audio-based wildfire detection system.
Biomedical decision-making involves integrating heterogeneous, high-dimensional data under cognitive, temporal, and computational constraints, yet presenting all available information can degrade decision quality for clinicians, LLMs, and statistical predictors alike. We introduce Dynamic Information Sub-Selection (DISS), an assistive framework that learns instance-specific feature subsets to forward to a fixed decision-maker under a limited interaction budget, improving decisions without modifying the underlying predictor. Evaluated across LLM-based biomedical reasoning, simulated cognitively-limited decision-makers, and clinician-assisted colorectal cancer risk assessment, DISS consistently improves decision quality through adaptive information curation.