Affiliations: [a] School of Foreign Language, Northeast Normal University, Changchun, Jilin 130000, China | [b] Dongan Experimental School Attached to Northeast Normal University, Changchun, Jilin 130000, China | [c] Panjin Liaodongwan Experimental High School, Panjin, Liaoning 124000, China
Correspondence:
[*]
Corresponding author: Jue Wang, School of Foreign Language, Northeast Normal University, Changchun, Jilin 130000, China. E-mail: [email protected].
Abstract: In order to make the swarm intelligence optimization learning method better serve the present agriculture, aiming at the complexity of agricultural production problems, a three-dimensional chaotic Drosophila Optimization Stochastic Forest prediction model is proposed. Firstly, the original Drosophila optimization algorithm is extended from two-dimensional search space to three-dimensional space, and chaos theory is introduced to initialize the population to avoid falling into local optimum. An improved three-dimensional chaotic Drosophila optimization algorithm is proposed. The experimental results show that the proposed method not only has better solution quality, but also has faster convergence speed. Then, the algorithm is introduced into the Stochastic Forest model, and the three-dimensional chaotic Drosophila optimization algorithm is used to train the stochastic forest to establish the optimal calculation model. Finally, the method is tested on rice pest data set. The experimental results show that the model has better prediction accuracy and can more effectively realize the prediction of rice pests.
