Research on distance from point in to hyperplane in euclidean space 歐氏空間中點(diǎn)到超平面的距離研究
The separating hyperplane of traditional support vector machines is sensitive to noises and outliers 摘要傳統(tǒng)的支持向量機(jī)分類超平面對(duì)噪聲和野值非常敏感。
When traditional support vector machines separate data containing noises , the obtained hyperplane is not an optimal one 使用傳統(tǒng)的支持向量機(jī)對(duì)含有噪聲的數(shù)據(jù)分類時(shí),所得到的超平面往往不是最優(yōu)超平面。
Svm maps input vectors nonlinearly into a high dimensional feature space and constructs the optimum separating hyperplane in the spade to realize modulation recognition 支撐矢量機(jī)把各個(gè)識(shí)別特征映射到一個(gè)高維空間,并在高維空間中構(gòu)造最優(yōu)識(shí)別超平面分類數(shù)據(jù),實(shí)現(xiàn)通信信號(hào)的調(diào)制識(shí)別。
The multiple - hyperplane classifier , which is investigated from the complexity of optimization problem and the generalization performance , is the explicit extension of the optimal separating hyperplanes classifier 多超平面分類器從優(yōu)化問(wèn)題的復(fù)雜度和運(yùn)行泛化能力兩方面進(jìn)行研究,是最優(yōu)分離超平面分類器一種顯而易見的擴(kuò)展。