Artificial Intelligence Prediction of Disease Vector Abundance Near Livestock Farms

Abstract

Vector-borne diseases associated with livestock are a significant issue when it comes to livestock health management due to their impact on productivity, pathogen dissemination, and economic sustainability of the agricultural system. Vectors like mosquitoes, ticks, and biting flies cause disease outbreaks through interactions with the livestock population and the environment they exist in. Traditional vector surveillance methods rely on manual sampling, regular field studies, and observations, which may be time-consuming and inefficient for the continuous evaluation of disease risk. The aim of this study is to develop an AI-based prediction system for determining vector abundance around livestock farms using environmental and farm-specific features. The suggested approach includes important variables such as temperature, humidity, rainfall, vegetation state, presence of water, livestock population density, and farm-specific characteristics. Several machine learning algorithms such as Support Vector Machine, Random Forest, Artificial Neural Network, and Gradient Boosting were tested to determine which algorithm is more efficient for prediction of vector abundance. According to the experimental results, it is possible to say that AI algorithms allow establishing links between the ecological environment and vector population and making predictions about high-risk periods and areas. This methodology is effective in supporting proactive disease management among livestock through an intelligent decision support system for early vector surveillance and preventive control. This study has made a significant contribution towards sustainable livestock disease management through an integration of artificial intelligence and environmental monitoring.