Crop Yield Prediction: Data Structure and Ai-Powered Methods

Smart farming, also known as intelligent agriculture, represents a modern stage in the development of agricultural science and practice. Its defining feature lies in the active application of artificial intelligence methods, particularly machine learning and deep learning, to address specific tasks aimed at ensuring sustainable crop production. (Research purpose) The aim of this study is to analyze data structures and compare machine learning and deep learning algorithms used in used in crop yield prediction. (Materials and methods) Using a convergent approach and applying methods of cognitive and semantic analysis, the authors examined the subject area of artificial intelligence applications in crop yield prediction. The study also explores key aspects related to the structure of input data, the main stages of implementing predictive models, and the most widely used machine learning and deep learning methods. (Results and discussion) The study presents the core data structure and methods for data acquisition, along with a typical workflow for implementing predictive analytics models for crop yield prediction. The most commonly used machine learning and deep learning methods are identified and their functional characteristics are examined in detail. Comparative analysis demonstrates that deep learning and hybrid approaches outperform traditional machine learning methods in terms of prediction accuracy, as measured by standard error metrics. (Conclusions) The findings confirm the advantages of deep learning methods (mean R² = 0.85) and hybrid approaches (mean R² = 0.87) in crop yield prediction under varying conditions and management interventions. Future research may focus on adapting modern AI approaches to spatial land use objects and crop types, with an emphasis on remote sensing data.

1. Bali N., Singla A. Emerging Trends in Machine Learning to Predict Crop Yield and Study Its Influential Factors: A Survey. Archives of Computational Methods in Enginee­ring. 2022. 29. 95-112 (In English). DOI: 10.1007/s11831-021-09569-8.

2. Shingade S.D., Mudhalwadkar R.P. Analysis of crop prediction models using data analytics and ML techniques: a review. Multimedia Tools and Applications. 2024. 83. 37813-37838 (In English). DOI: 10.1007/s11042-023-170386.

3. Meghraoui K., Sebari I., Pilz J.et al/. Applied Deep Learning-Based Crop Yield Prediction: A Systematic Analysis of Current Developments and Potential Challenges. Technologies. 2024. 12. N4. 43 (In English). DOI: 10.3390/technologies12040043.

4. Stepanov A.S., Aseeva T.A., Dubrovina K.N. Building and precision assessment of regression models for determining of cereals’ and legumes’ crop yield based on Earth remote sensing data and climatic characteristics. InterCarto. InterGIS. 2020. 26. N3. 159-169 (In Russian). DOI: 10.35595/2414-9179-2020-3-26-159-169.

5. Strashnaya A.I., Bereza O.V., Klang P.S. Forecasting grain crop yield based on the integration of ground and satellite data in the subjects of the Southern Federal District. Hyd­rometeorological Research and Forecasting. 2021. N2 (380). 111-137 (In Russian). DOI: 10.37162/2618-9631-2021-2-111-137.

6. Bischokov R.M., Akhmatov M.M. Agricultural crops analysis and yield forecast by fuzzy logic methods. Polythematic online Scientific Journal of Kuban State Agrarian University. 2021. N168. 274-287 (In Russian). DOI: 10.21515/1990-4665-168-020.

7. Blagov D.A., Mitrofanov S.V., Nikitin V.S. et al. Mathematical models and software package for forecasting crop yields. Agrotekhnika i energoobespechenie. 2019. N3 (24). 182-188 (In Russian). EDN: OMWNXT.

8. Arkhipova M.Yu., Smirnov A.I. Current trends in crop yield forecasting based on the use of econometric models. Voprosy statistiki. 2020. 27. N5. 65-75 (In Russian). DOI: 10.34023/2313-6383-2020-27-5-65-75.

9. Momenpour S.E., Bazgeer S., Moghbel M. A bibliometric analysis of the literature on crop yield prediction: insights from previous findings and prospects for future research. International Journal of Biometeorology. 2024. N68. 829-842 (In English). DOI: 10.1007/s00484-024-02628-2.

10. Makeev K.A., Grecheneva A.V., Kotov Ya.S. et al. Comparison of the effectiveness of machine learning algorithms in crop yield forecasting tasks. Izvestiya Tula State University. 2023. N2. 203-206 (In Russian). DOI: 10.24412/2071-6168-2023-2-203-206.

11. Dudko Yu.V., Marinkin E.B., Vladimirova T.M. Application of neural network technologies in predicting soil fertility with leguminous crops and predicting their yields. Voprosy Nauki: Innovatika, Tekhnika i Tekhnologii. 2019. N1. 67-72 (In Russian). EDN: YXPXCX.

12. Rogachev A.F. System analysis and forecasting of time seriesof crop production based on autocorellation functionsand neural network technologies. Proceedings of Lower Volga Agro-University Complex. 2018. N3 (51). 309-316 (In Russian).EDN: VOQSWI.

13. Rogachev A.F., Melikhova E.V. Justification of algorithms and tools for neural networkforecasting of agricultural productivity using retrospective data. Proceedings of Lo­wer Volga Agro-University Complex. 2020. N1(57). 290-302 (In Russian). DOI: 10.32786/2071-9485- 2020-01-29.

14. Bischokov R.M. Analysis, modelling and forecasting of crop yields using artificial neural networks. RUDN Journal of Agronomy and Animal Industries. 2022. 17. N2. 146157 (In Russian). DOI: 10.22363/2312-797X-2022-17-2-146-157.

15. Boppudi S.JS. Deep ensemble model with hybrid intelligence technique for crop yield prediction. Multimedia Tools and Applications. 2024. 1-21 (In English). DOI: 10.1007/s11042-024-18354-1.

16. Oikonomidis A., Catal C., Kassahun A. Hybrid deep learning-based models for crop yield prediction. Applied Artificial Intelligence. 2022. 36. N1. 2031822. (In English). DOI: 10.1080/08839514.2022.2031823.

17. Cao J., Wang H., Li J. et al. Improving the forecasting of winter wheat yields in Northern China with machine learning–dynamical hybrid subseasonal-to-seasonal ensemble prediction. Remote Sensing. 2022. N14(7). 1707 (In English). DOI: 10.3390/rs14071707.

18. Cedric L.S., Adoni W.Y.H., Aworka R. et al. Crops yield prediction based on machine learning models: Case of West African countries. Smart Agricultural Technology. 2022. 2. 100049 (In English). DOI: 10.1016/j.atech.2022.100049.

19. Elavarasan D., Vincent D.R., Sharma V. et al. Forecasting yield by integrating agrarian factors and machine learning models: A survey. Computers and Electronics in Agriculture. 2018. N155. 257-282 (In English). DOI: 10.1016/j.compag.2018.10.024.

20. Meghraoui K., Sebari I., Pilz J. et al. Applied deep learning-based crop yield prediction: a systematic analysis of current developments and potential challenges. Techno­logies. 2024. N12(4). 43. (In English). DOI: 10.3390/technologies12040043.