引入激活扩散的类分布关系近邻分类器  

Introducing Class-Distribution Relational Neighbor Classifierwith Activation Spreading

作　　者：董飒[1,2] 欧阳若川[3] 徐海啸[1] 刘杰[1,2] 刘大有[1,2] 李婷婷 王鑫禄[1,4] DONG Sa;OUYANG Ruochuan;XU Haixiao;LIU Jie;LIU Dayou;LI Tingting;WANG Xinlu(College of Computer Science and Technology,Jilin University,Changchun 130012,China;Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education,Jilin University,Changchun 130012,China;Faculty Work Department of Party Committee,Jilin University,Changchun 130012,China;College of International Education,Jilin University,Changchun 130012,China)

机构地区：[1]吉林大学计算机科学与技术学院,长春130012 [2]吉林大学符号计算与知识工程教育部重点实验室,长春130012 [3]吉林大学党委教师工作部,长春130012 [4]吉林大学国际教育学院,长春130012

出　　处：《吉林大学学报（理学版）》2024年第4期915-922,共8页Journal of Jilin University:Science Edition

基　　金：国家自然科学基金(批准号:61502198)。

摘　　要：针对同质性关系分类器基于一阶Markov假设简化处理的局限性,在类分布关系近邻分类器构建类向量和参考向量时,引入局部图排序激活扩散方法,并结合松弛标注的协作推理方法,通过适当扩大分类时邻居节点的范围增加网络数据中待分类节点的同质性,从而降低分类错误率.对比实验结果表明,该方法扩大了待分类节点的邻域,在网络数据上分类精度较好.Aiming at the limitation of the simplifying the processing of homophily relational classifiers based on first-order Markov assumption,when constructing the class vector and reference vector in the class-distribution relational neighbor classifier,we introduced the activation spreading algorithm of local graph ranking,combined with the relaxation labeling collective inference method.By appropriately expanding the range of neighboring nodes during classification,we increased the homophily of nodes to be classified in network data,thereby reducing the error rate of classification.The comparative experimental results show that this method expands the neighborhood of nodes to be classified,and has good classification accuracy on network data.

关 键 词：人工智能 网络数据分类 激活扩散 类分布关系近邻分类器 协作推理 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]