AGRI-SMART HUB: A Multilingual Integrated Platform for Comprehensive Farming Solutions and Support

This research work proposes an integrated web platform that uses machine learning and web development to solve pressing challenges in agriculture. While the world population increases exponentially and the land for farming is constant, deforestation and climate change lower crop yields, posing a threat to food security. For a growing population's demand, we need to shift from conventional farming to a technological approach. Nowadays everybody has access to phones and the internet. Our website aims to strengthen farmers by providing them with new resources and government schemes for enhanced productivity and sustainability. This study integrates the multiple machine learning models that fulfill the farmers' needs. These include a crop recommendation system that uses parameters such as rainfall, humidity, soil pH, and moisture to suggest suitable crops. our crop yield prediction model that analyzes rainfall and NPK (nitrogen, phosphorus, potassium) levels to forecast yields; and a transfer learning-based CNN model that takes the leaf as input and detects the disease in the early stage. Moreover, we use Retrieval-Augmented Generation (RAG) to assist farmers so that they can use the government's issues pdf regarding agriculture and ask questions based on that pdf, which they did not understand in their regional language. This study highlights the potential of technology-based solutions in revolutionizing agriculture and responding to the global issue of food security in an increasingly dynamic world.