单调分类问题是特征与类别之间带有单调性约束的有序分类问题.对于符号数据的单调分类问题已有较好的方法,但对于数值数据,现有的方法分类精度和运行效率有限.提出一种基于决策森林的单调分类方法(monotonic classification method base...单调分类问题是特征与类别之间带有单调性约束的有序分类问题.对于符号数据的单调分类问题已有较好的方法,但对于数值数据,现有的方法分类精度和运行效率有限.提出一种基于决策森林的单调分类方法(monotonic classification method based on decision forest,MCDF),设计采样策略来构造决策树,可以保持数据子集与原数据集分布一致,并通过样本权重避免非单调数据的影响,在保持较高分类精度的同时有效提高了运行效率,同时这种策略可以自动确定决策森林中决策树的个数.在决策森林进行分类时,给出了决策冲突时的解决方法.提出的方法既可以处理符号数据,也可以处理数值数据.在人造数据集、UCI及真实数据集上的实验数据表明:该方法可以提高单调分类性能和运行效率,缩短分类规则的长度,解决数据集规模较大的单调分类问题.展开更多
By using the continuation theorem of coincidence degree theory, sufficient conditions are obtained for the existence of positive periodic solutions of a delayed predator prey system with nonmonotonic functional respon...By using the continuation theorem of coincidence degree theory, sufficient conditions are obtained for the existence of positive periodic solutions of a delayed predator prey system with nonmonotonic functional response in a periodic environment.展开更多
This paper deals with a new class of nonlinear set valued implicit variational inclusion problems involving （A, η）-monotone mappings in 2-uniformly smooth Banach spaces. Semi-inner product structure has been used t...This paper deals with a new class of nonlinear set valued implicit variational inclusion problems involving （A, η）-monotone mappings in 2-uniformly smooth Banach spaces. Semi-inner product structure has been used to study the （A, η）-monotonicity. Using the generalized resolvent operator technique and the semi-inner product structure, the approximation solvability of the proposed problem is investigated. An iterative algorithm is constructed to approximate the solution of the problem. Convergence analysis of the proposed algorithm is investigated. Similar results are also investigated for variational inclusion problems involving （H, η）-monotone mappings.展开更多
基金Supported by the National Science Foundation of China(11302002)the National Science Research Project of Anhui Educational Department(KJ2012Z127)the PhD research startup foundation of Anhui Normal University(151203)
文摘单调分类问题是特征与类别之间带有单调性约束的有序分类问题.对于符号数据的单调分类问题已有较好的方法,但对于数值数据,现有的方法分类精度和运行效率有限.提出一种基于决策森林的单调分类方法(monotonic classification method based on decision forest,MCDF),设计采样策略来构造决策树,可以保持数据子集与原数据集分布一致,并通过样本权重避免非单调数据的影响,在保持较高分类精度的同时有效提高了运行效率,同时这种策略可以自动确定决策森林中决策树的个数.在决策森林进行分类时,给出了决策冲突时的解决方法.提出的方法既可以处理符号数据,也可以处理数值数据.在人造数据集、UCI及真实数据集上的实验数据表明:该方法可以提高单调分类性能和运行效率,缩短分类规则的长度,解决数据集规模较大的单调分类问题.
文摘By using the continuation theorem of coincidence degree theory, sufficient conditions are obtained for the existence of positive periodic solutions of a delayed predator prey system with nonmonotonic functional response in a periodic environment.
文摘This paper deals with a new class of nonlinear set valued implicit variational inclusion problems involving （A, η）-monotone mappings in 2-uniformly smooth Banach spaces. Semi-inner product structure has been used to study the （A, η）-monotonicity. Using the generalized resolvent operator technique and the semi-inner product structure, the approximation solvability of the proposed problem is investigated. An iterative algorithm is constructed to approximate the solution of the problem. Convergence analysis of the proposed algorithm is investigated. Similar results are also investigated for variational inclusion problems involving （H, η）-monotone mappings.
