| 基于优化Logistic分类算法的山体滑坡研究 |
 点此下载全文 |
| 引用本文：刁君华1,2，冯向萍1，马新春2.基于优化Logistic分类算法的山体滑坡研究[J].计算技术与自动化,2025,(4):21-25 |
| 摘要点击次数: 101 |
| 全文下载次数: 63 |
|
|
| 中文摘要:滑坡是一种常见的地质灾害，受到滑坡曲率、坡度等因素的影响。然而这些因素一般是以连续型数据存在，传统的机器学习分类算法无法适应此种数据类型，一般采用人工处理方式将其转换为离散型数据，这种方式不但耗费人力，而且无法准确划分影响因素离散化后的区间范围，产生噪声。针对于此实验提出了一种集成机器学习模型，以Logistic算法为分类器，以随机森林算法为连续性数据处理工具，减少人工处理方式产生的噪声影响。模型在滑坡灾害数据集上得以验证，F1值达到了90.1%。 |
| 中文关键词:机器学习 人工处理 噪声 Logistic 随机森林 |
| |
| Research on Landslides Based on Optimized Logistic Classification Algorithm |
|
|
| Abstract:Landslide is a common geological disaster, which is affected by factors such as landslide curvature and slope. However, these factors generally exist in the form of continuous data. Traditional machine learning classification algorithms cannot adapt to this type of data. Manual processing is generally used to convert it into discrete data. This method is not only labor-intensive, but also cannot accurately classify the influencing factors. The discretized interval range generates noise. For this experiment, an integrated machine learning model was proposed, using the Logistic algorithm as the classifier and the random forest algorithm as the continuous data processing tool to reduce the noise impact caused by manual processing. The model was verified on the landslide disaster data set, and the F1 value reached 90.1%. |
| keywords:machine learning manual processing noise Logistic random forest |
| 查看全文 查看/发表评论 下载pdf阅读器 |
位访问者,今日访问
