APPLICATION OF MACHINE LEARNING ALGORITHMS IN MANAGING HUMAN WILDLIFE CONFLICT TO BOOST RICE YIELD IN RICE PLANTATION FARMS IN ANAMBRA STATE

Rice farming in Anambra State, Nigeria, is a major source of food security and economic stability, producing about 300,000 metric tonnes annually across 100,000 hectares. However, human-wildlife conflict (HWC) poses a significant threat to productivity, as birds, rodents, primates, ungulates, and stray cattle invade farmlands, consuming grains and damaging crops. Traditional methods such as manual guarding and scare tactics have proven labour-intensive, costly, and largely ineffective. This study investigates the effectiveness of Artificial Intelligence (AI) and Machine Learning (ML) interventions in mitigating HWC and enhancing rice production. A mixed-method approach was adopted, involving 65 respondents, including managers, assistant managers, and operational staff from Coscharis Farm, Josan Rice Mills, Omor Rice Farms, Easter plains integrated Farms and Stine industries Nigeria limited. AI-driven interventions implemented included flashing light systems, movement identifiers, call identifiers, and image recognition with alarm systems. Data were analyzed using Analysis of Variance (ANOVA) in SPSS version 25 to assess the effectiveness of these technologies. The results show that all interventions significantly reduced wildlife intrusions, protected crops, and improved rice yields, with image recognition and alarm systems being the most effective. The study concludes that AI and ML technologies provide scalable, cost-effective, and environmentally sustainable solutions to HWC, promoting food security and sustainable agriculture in Anambra State