A Comprehensive Agricultural Dataset designed for XAI in Crop Data-Driven Decision Making for the Indian Population

This dataset presents a comprehensive agricultural data repository developed to support crop prediction, yield estimation, fertilizer analysis, and data-driven agricultural decision-making across the Indian agricultural sector. The dataset integrates verified governmental and institutional agricultural data sources into a unified, machine learning-ready tabular dataset. It combines climate, soil, fertilizer consumption, irrigation methods, crop production, and yield information to support agricultural analytics, explainable artificial intelligence (XAI), precision farming, and sustainable agriculture research. Dataset Characteristics: Total Records: 1,194,806 Total Features: 26 Geographic Coverage: All 28 Indian States Temporal Coverage: 1980 onwards Data Formats: CSV and XLSX Features Included: Temperature Humidity Rainfall State and District Information Soil Type Nitrogen Consumption Phosphate Consumption Potash Consumption NPK Consumption Statistics Fertilizer Consumption per Hectare Total Cultivated Area Crop Production Crop Yield Crop Name Cropping Season Intercropping Information Crop Rotation Practices Irrigation Methods Chemical Fertilizer Usage Organic Fertilizer Usage Agricultural Benefits and Management Information The dataset was constructed through multi-source integration, preprocessing, data cleaning, text standardization, unit normalization, composite key-based merging, and missing-value handling to ensure consistency, reliability, and compatibility with machine learning and data analytics workflows. Potential Applications: Crop recommendation systems Crop yield prediction Agricultural production forecasting Fertilizer recommendation systems Precision agriculture Explainable Artificial Intelligence (XAI) Agricultural analytics Climate-aware crop planning Sustainable agriculture research Related Dataset Groundwater data has been published separately to facilitate independent analysis of groundwater resources and irrigation studies while maintaining a modular dataset structure. *This dataset is intended strictly for academic and research purposes.*