Treffer: Crop Prediction Based on Environment Variables using Data Mining Technologies.

Purpose: Sustainable agriculture is essential for addressing food security challenges and enhancing the socio-economic status of farmers. Integrating modern technologies with the agricultural sector is a key solution to overcome many issues. Therefore, this study aimed to apply data mining technologies to identify the most suitable crop types for specific land based on factors: weather conditions (rainfall, temperature, humidity), crop prices, and soil conditions of lands. Potato, tomato, green gram, and red onions are crop types. The soil conditions are determined based on the locations specified as Grama Niladari Divisions of the Badulla District. Research Method: The Cross-Industry Standard Processfor Data Mining (CRISP-DM) methodology was applied to develop the recommended system. The MYSQL database, WEKA libraries (is it "and WEKA libraries" - please check with the researcher), which integrated the Java programming language, were used to implement the prediction models. The prediction quality of models was evaluated using different evaluation metrics. Findings: The Random Forest tree and Random tree outperformed the other models in predicting the weather conditions for the next four months from the system usage date in a weekly manner and the variability in crop prices, respectively. Once the location is selected, the suitable crops are ranked based on the prevailing soil conditions in that location. Then, the most suitable crops were selected based on the predicted weather factors, which best fit with the ranked crops, together with the predicted prices. Value: The system ranks suitable crops for specific lands. These findings assistfarmers and decision-makers in transforming agriculture into a profitable, economically stable industry. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Agricultural Sciences (Sri Lanka) is the property of Sabaragamuwa University of Sri Lanka, Faculty of Agricultural Sciences and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)