Property taxes, notably land rent, are crucial revenue sources particularly in third-world countries. Land rent is determined by factors like location and market forces and is governed in Kenya by the Land Act No. 6 of 2012. Manual valuation methods have led to inconsistencies, inefficiencies, and corruption. Many countries have adopted the Automated Valuation System (AVS), which lowers costs, enhances efficiency, ensures transparency, and boosts revenue collection. However, the adoption of AVS in developing nations remains largely uncharted in literature. The study seeks to fill this gap by scrutinizing various Automated Valuation Models (AVMs) for land rent taxation in Nairobi City County using data from the years 2020 to 2023. These datasets l existing land rents, property values, zoning ordinances, and public facilities’ locations. The research will use a case study of Nairobi City County, utilizing a cross-sectional quantitative research design and stratified random sampling considering the 20 zones in Nairobi as strata. The study will employ tools like ArcGIS Pro and R Statistic, the research will utilize techniques from multiple regression to Artificial Neural Networks to create prediction models for land rent. After assessing each model's accuracy, the most effective will be recommended for Kenya's use.