The SVM method is based on seeking on the Structural Risk Minimization by few learning samples supporting, and it has important feature such as good generalization and classification performance, etc.
支持向量机方法基于小学习样本条件下,通过寻求结构风险最小,以期获得良好的分类效果和泛化能力。
Since the data samples in machine learning and pattern recognition problems generally distribute in multi-modal distribution, this thesis proposed a prototype based feature ranking model.
由于模式识别、机器学习等问题的复杂性比较高,数据分布通常呈现多模态分布。
This was further demonstrated with the success of their computer learning models in being able to identify each participant based solely on their samples.
这进一步证实了他们仅凭参与者样本就能识别出每个参与者的计算机学习模型是成功的。
应用推荐