ANNOUNCEMENTS
The financial landscape has undergone significant transformations due to technological advancements, regulatory changes, and evolving market dynamics. Financial analysis, which once relied on traditional methods, now leverages advanced algorithms and artificial intelligence, with spatial finance becoming increasingly important by integrating geospatial data into financial theory. Given agriculture's critical role in India's economy and the challenges it faces, such as fragmented landholdings, water scarcity, and climate change, precise land price estimation is essential for informed decision-making by farmers, policymakers, stakeholders, and financial institutions.
This paper explores the application of spatial finance in agriculture, focusing on its potential to enhance the estimation of agricultural land prices in India. To address the limitations of traditional hedonic pricing models, the study evaluates both market and environmental variables to assess the performance of modern machine learning algorithms, specifically XGBoost, Support Vector Machines (SVM), and Random Forest. Additionally, it employs K-Fold cross-validation to ensure accuracy, as these algorithms capture complex, non-linear relationships between land prices and various influencing factors, providing a more nuanced approach to land valuation.
Concentrating on Uttar Pradesh, a key agricultural state in India, the study highlights the lack of an open-source platform for land price information accessible to grassroots communities and other stakeholders. As a final product, the research develops a dynamic web application incorporating the predictive model, enabling users to visualize land assessments and access valuable agricultural insights. By addressing the knowledge gap in agricultural land price estimation, this research aims to promote sustainable agriculture and democratise data access in India.
Keywords: Spatial Finance; Agriculture; Land price estimation; XGBoost; Random Forest; SVM; Web application.
