Salybekova Nurdana, Issayev Gani, Serzhanova Aikerim, Mikhailov Valery
Department of Biology, Khoja Akhmet Yassawi International Kazakh-Turkish University, Turkistan, Kazakhstan.
Department of System Analysis and Information Technologies, Kazan Privolzhsky Federal University, Kazan, Russian Federation.
Bot Stud. 2024 Dec 19;65(1):39. doi: 10.1186/s40529-024-00445-9.
The research aims to assess the effectiveness of artificial intelligence models in predicting the risk level in tulip greenhouses using different varieties. The study was conducted in 2022 in the Almaty region, Panfilov village.
Two groups of 10 greenhouses each (area 200 m2) were compared: the control group used standard monitoring methods, while the experimental group employed AI-based monitoring. We applied ANOVA, regression analysis, Bootstrap, and correlation analysis to evaluate the impact of factors on the risk level. The results demonstrate a statistically significant reduction in the risk level in the experimental group, where artificial intelligence models were employed, especially the recurrent neural network "Expert-Pro." A comparison of different tulip varieties revealed differences in their susceptibility to risks. The results provide an opportunity for more effective risk management in greenhouse cultivation.
The high accuracy and sensitivity exhibited by the "Expert-Pro" model underscore its potential to enhance the productivity and resilience of crops. The research findings justify the theoretical significance of applying artificial intelligence in agriculture and its practical applicability for improving risk management efficiency in greenhouse cultivation conditions.
本研究旨在评估人工智能模型在预测不同品种郁金香温室风险水平方面的有效性。该研究于2022年在阿拉木图地区的潘菲洛夫村进行。
比较了两组各10个温室（面积200平方米）：对照组采用标准监测方法，而实验组采用基于人工智能的监测。我们应用方差分析、回归分析、自助法和相关分析来评估各因素对风险水平的影响。结果表明，在采用人工智能模型的实验组中，风险水平有统计学意义的降低，尤其是循环神经网络“专家-Pro”。不同郁金香品种的比较显示出它们对风险的易感性存在差异。研究结果为温室种植中更有效的风险管理提供了契机。
“专家-Pro”模型表现出的高精度和敏感性凸显了其提高作物生产力和抗逆性的潜力。研究结果证明了人工智能在农业中的理论意义及其在提高温室种植条件下风险管理效率方面的实际适用性。
