基于特征选择的SSA-XGBoost水质量预测研究

Research on SSA-XGBoost Water Quality Prediction Based on Feature Selection

摘要: 为了能够更好的实现水资源的利用，针对目前对水质预测研究中存在的特征参数复杂、单一模型预测模型精度和适应度欠佳等问题，提出了一种基于XGBoost的水质预测模型。首先利用主成分分析方法对特征进行选择，降低问题复杂度和计算成本，并对数据中的缺失值进行填充，其次采用麻雀搜索算法(SSA)对XGBoost模型中的参数进行优化，采用优化后的参数对水质进行预测。最后在不同实验条件下对水质进行预测，实验结果证明，本文提出的SSA-XGBoost方法与现有方法相比，具有更优秀的性能。

Abstract: In order to better realize the utilization of water resources, a water quality prediction model based on XGBoost is proposed in view of the problems existing in the current research on water quality prediction, such as complex characteristic parameters, poor precision and fitness of a single model prediction model, etc. Firstly, the principal component analysis method is used to select features, reduce problem complexity and computational costs, and fill in missing values in the data. Secondly, the sparrow search algorithm (SSA) is used to optimize the parameters in the XGBoost model, and the optimized parameters are used to predict water quality. Finally, water quality was predicted under different experimental conditions, and the experimental results showed that the SSA-XGBoost method proposed in this paper has better performance compared to existing methods.