Price Calculation for Cashew using supervised learning techniques

Inconsistency and subjectivity in grading cashews result in price inefficiencies, market conflicts, and a lack of trust in agricultural value chains. To address this, this paper suggests a grading and price prediction framework for cashews using artificial intelligence, comprising machine learning, deep learning, and an agentic optimization component. The structured data set comprising 8,000 annotated cashew samples was constructed using official grading standards, considering visual attributes and contextual metadata such as size, quality, location, date, and market price. To classify cashew quality, this paper implemented a Random Forest classifier and a Support Vector Machine classifier as a baseline, and a neural network to predict visual quality. The proposed framework links visual quality assessment with economic intelligence, making it a potential tool for modernizing cashew trade, eliminating human bias, and promoting price transparency. The framework has wider implications for other quality-based agricultural commodities.