摘要
为提高气体传感器检测数据分类的准确度,提出一种随机森林算法,利用传感器响应输出电压和响应时间构建模型的特征向量,实验研究二氧化碳气体体积分数的分类,分析决策树数量对模型分类准确度的影响,验证模型分类的有效性。结果表明,与传统分类方法相比,随机森林算法模型的准确度能够达到94.6%,提高了分类准确度和计算效率。该研究为气体传感器数据分析提供了参考。
This paper proposes a random forest algorithm to improve the accuracy of gas sensor detection data classification.The study works by constructing the feature vector of the model by using output voltage and response time of the sensor response;experimenting and studying the classification of carbon dioxide concentration;analyzing the influence of the number of decision trees on the classification accuracy of the model and verifying the effectiveness of the model classification.The results show that compared with the traditional classification method,the accuracy of the random forest algorithm model can reach 94.6%,which improves the classification accuracy and calculation efficiency.This study provides a reference for gas sensor data analysis.
作者
沈斌
靳春博
刘新蕾
Shen Bin;Jin Chunbo;Liu Xinlei(School of Safety Engineering,Heilongjiang University of Science&Technology,Harbin 150022,China)
出处
《黑龙江科技大学学报》
2022年第6期711-715,共5页
Journal of Heilongjiang University of Science And Technology
基金
国家自然科学基金项目(52074111)
黑龙江省自然科学基金项目(YQ2020E034)。
关键词
随机森林算法
数据分类
特征值向量
random forest algorithm
data classification
eigenvalue vector
