摘要
逻辑回归(LR)作为监督学习的二元分类广义线性分类器,在处理线性数据方面表现出结构简单、解释性强,拟合效果好的特点。然而,当面对高维、不确定性和线性不可分数据时,逻辑回归的分类效果受到限制。针对逻辑回归的固有缺陷,引入粒计算理论,借助粒化的优势提出一种新型的逻辑回归模型:旋转粒逻辑回归。通过引入旋转粒化理论,在特征两两组合形成的平面坐标系上旋转不同角度,构建旋转粒子,多平面坐标系上粒化构造旋转粒向量。进一步定义粒的大小、度量和运算规则,提出旋转粒逻辑回归的损失函数。通过求解损失函数,得到旋转粒逻辑回归的优化解。最后,采用多个UCI数据集进行实验,从多个评价指标比较的结果表明旋转粒逻辑回归模型的有效性。
LR serves as a generalized linear classifier for binary classification in supervised learning,exhibiting characteristics of simplicity in structure,strong interpretability,and effective fitting when dealing with linear data.However,its classification performance becomes limited when confronted with high-dimensional,uncertain,and linearly inseparable data.To address the inherent limitations of logistic regression,this paper introduced the theory of granular computing and proposed a novel logistic regression model called rotating granular logistic regression(RGLR).This paper introduced the theory of rotating granulation,where different angles of rotation were applied to pairs of features forming a plane coordinate system.This process constructed rotating granules by rotating pairs of features at various angles on the plane coordinate system,and granulated to form rotating granule vectors on multiple plane coordinate systems.This paper further defined the size,measurement,and operational rules of granules,and proposed a loss function for rotating granular logistic regression.The optimized solution of the rotating granular logistic regression was obtained by solving the value of the loss function.Finally,experiments are conducted using multiple UCI datasets,and the results compared across various evaluation metrics,indicate the effectiveness of the rotating granular logistic regression model.
作者
孔丽茹
陈玉明
傅兴宇
江海亮
许进程
Kong Liru;Chen Yuming;Fu Xingyu;Jiang Hailiang;Xu Jincheng(College of Computer&Information Engineering,Xiamen University of Technology,Xiamen Fujian 361024,China;Xiamen Wanyin Intelligent Technology Co.,Ltd.,Xiamen Fujian 361024,China)
出处
《计算机应用研究》
CSCD
北大核心
2024年第8期2398-2403,共6页
Application Research of Computers
基金
国家自然科学基金资助项目(61976183)。
关键词
逻辑回归
粒计算
向量旋转
粒逻辑回归
损失函数
logistic regression
granular computing
rotating vector
rotating granular logistic regression
loss functions
作者简介
通信作者:孔丽茹(1998-),女,山东济宁人,硕士研究生,主要研究方向为机器学习、粒计算(k599651599@163.com);陈玉明(1977-),男,江西人,教授,博导,博士,主要研究方向为人工智能、粒计算;傅兴宇(1997-),男,福建三明人,硕士研究生,主要研究方向为粒计算;江海亮(1997-),男,福建龙岩人,硕士研究生,主要研究方向为粒计算;许进程(1982-),男,福建厦门人,工程师,硕士,主要研究方向为大数据.
