Intuitionistic fuzzy based machine learning models for prediction of the oilseed prices

Oilseed prices are inherently volatile and uncertain, making accurate predictions is important for the stakeholders. In time series forecasting, fuzzy techniques have proven effective for managing complex and uncertain datasets. This study introduces an innovative approach to predicting oilseeds prices by developing intuitionistic fuzzy based machine learning models. The model integrates intuitionistic fuzzy logic with stochastic and advanced machine learning techniques to enhance predictive accuracy. The main objective is to assess how this integration improves prediction accuracy, focusing on monthly wholesale prices of Sunflower from various markets in Karnataka, covering the period from January 2010 to June 2024 from the AGMARKNET portal (https://agmarknet.gov.in/). Comparative analysis with traditional models demonstrated the superior performance of the intuitionistic fuzzy based models, particularly in reducing prediction errors and accurately capturing market trends. This research underscores the potential of integrating fuzzy logic into machine learning frameworks, offering a valuable tool for stakeholders in agricultural economics and commodity trading.

⓿ A study investigated the impact of using fuzzy logic to manage uncertainty.

⓿ A new method for combining machine learning (ML) models with intuitionistic fuzzy logic is proposed.

⓿ A comparison between fuzzy-based ML and traditional models is conducted.

⓿ The prediction of monthly wholesale sunflower prices shows significant improvement.