The Implementation of Machine Learning and Deep Learning Algorithms for Crop Yield Prediction in Agriculture

5. Conclusion
In recent years, machine learning techniques, 
such as multivariate linear regression (MLR), 
multiple regression (MR), and deep neural networks 
(DNN), have shown promising results in crop yield 
prediction. In this paper, we evaluated the 
performance of MLR, MR, and DNN models in 
predicting crop yield using a publicly available 
dataset. Our results show that GBRT outperforms 
MLR,DNN and MR models in terms of prediction 
accuracy, with lower mean absolute error (MAE) 
and root mean squared error (RMSE) values. This 
indicates that GBRT is better suited for modeling 
multi-functional relationships between crop yield 
and various environmental and management factors.
However, we also note that the choice of 
algorithm for crop yield prediction depends on 
several factors, including the complexity of the 
problem, the amount and quality of data, and the 
specific application requirements. While GBRT 
may perform better in some cases, MLR, DNN and 
MR models can be more interpretable and easier to 
implement in certain scenarios. 
In our future works, we aim to expand the 
dataset by collecting more data with varying 
specifications, including a new features for effecting 
crops, creating new application to collect the 
agricultural data from farmers, reducing range of 

---

Bulletin of TUIT: Management and Communication Technologies
Nodir Rahimov, Dilmurod Khasanov 
2023.Vol-2(4) 
learning area (specific area from central Asia). By 
doing so, we can improve the generalization and 
prediction performance of the prediction model, 
making it more effective in the real world. 
References
1. Lee,W.; Jung, T.-Y.; Lee, S. Dynamic 
Characteristics Prediction Model for Diesel 
Engine Valve Train Design Parameters Based 
on Deep Learning. Electronics 2023, 12, 
1806. 
https://doi.org/10.3390/electronics12081806 
2. Amit Kumar Srivastava, Nima Safaei, Saeed 
Khaki, Gina Lopez, Wenzhi Zeng, Frank 
Ewert, Thomas Gaiser, Jaber Rahimi. Winter 
wheat yield prediction using convolutional 
neural networks from environmental and 
phenological data. 2022 
3. Koirala A, Walsh KB, Wang Z, McCarthy C. 
Deep learning–method overview and review 
of use for fruit detection and yield estimation. 
2019 
4. Dharani M, Thamilselvan R, Natesan P, 
Kalaivaani P, Santhoshkumar S. Review on 
crop prediction using deep learning 
techniques. 2021 
5. van Klompenburg T, Kassahun A, Catal C. 
Crop yield prediction using machine learning: 
a systematic literature review. 2020 
6. Alexandros Oikonomidis,Cagatay Catal, 
Ayalew Kassahuna. Deep learning for crop 
yield prediction: a systematic literature 
review. 2022 
7. Yifei Huang, Yuhua Liu, Chenhui, Changbo 
Wang. GBRTVis: online analysis of gradient 
boosting regression tree. 2018 
8. Huang Hui, Rong Jia, Xiaoyu Shi. Feature 
selection and hyper parameters optimization 
for short-term wind power forecast. 2021 
9. Chuan Lin, Qing Chang, Xianxu Li. A 
Deep Learning Approach for MIMO-NOMA 
Downlink Signal Detection. 2019 
10. Nie, P.; Roccotelli, M.; Fanti, M.P.; Ming, Z.; 
Li, Z. Prediction of home energy 
consumption based on gradient boosting 
regression tree. Energy Rep. 2021, 7, 1246–
1255. 
11. Jiang, S.; Li, J.; Zhang, S.; Gu, Q.; Lu, C.; 
Liu, H. Landslide risk prediction by using 
GBRT algorithm: Application of artificial 
intelligence in disaster prevention of energy 
mining. Process. Saf. Environ. Prot. 2022, 
166, 384–392. 
12. Saeed Khaki*, Lizhi Wang, Crop Yield 
Prediction Using Deep Neural Networks. 
2019 
13. N.Rahimov, D.Khasanov,“The application 
of multiple linear regression algorithm and 
python for crop yield prediction in 
agriculture”, Harvard educational and 
scientific review, Vol.2. Issue 1 Pg. 181-187. 
14. N.Rahimov, 
D.Khasanov,J.Kuvandikov, 
“Structural-funtional 
organization 
correctness of knowledge models of product 
systems”, Harvard educational and scientific 
review, Vol.2. Issue 2 Pg. 1-9. 
15. N.Rahimov, 
D.Khasanov, 
“The 
mathematical essence of logistic regression 
for machine learning”, International 
Journal of Contemporary Scientific and 
Technical Research. Pg. 102-105.
16. Hui, H. Rong, J. Xiaoyu, S. Jun,L. Jian, D.,
Feature selection and hyper parameters 
optimizationfor short-term wind power 
forecast. 
https://doi.org/10.1007/s10489-
021-02191-y