ANN model on paddy farmers’ climate change adaptation in Western Odisha

In Western Odisha, where paddy cultivation is the primary source of income, climate change poses serious issues to the agricultural sector. Using an Artificial Neural Network (ANN) model to analyze the relationship between paddy farmers' socio-economic characteristics and their adaptation strategies regarding climate change. The present study was conducted in 2024-25 to investigate the adaptation strategies of paddy farmers in the Balangir and Kalahandi districts of Western Odisha. Located in the KBK (Kalahandi-Bolangir-Koraput) region, these two districts are among the most impoverished in the state, where most of the residents are residing Below the Poverty Line (BPL). 180 farmers were randomly recruited from 30 villages, and their adaptation techniques were reviewed through structured interviews. Paddy farmers' adaptation tactics to climate change include high levels of agro-diversification (67.22%), low levels of varieties (41.11%), low levels of soil conservation measures (50%), and medium levels of land use management (66.11%). With the best hidden nodes—that is, 3,3 for the soil conservation measures component and 5,3 for the agro-diversification, varieties, and land use management component—an Artificial Neural Network (ANN) model was created. The optimal neural network architecture had a total minimum Mean Square Error (MSE) of 4.983.