小世界邻域优化的局部线性嵌入算法

通过分析稀疏数据或噪声数据,导出局部线性嵌入(LLE)算法出现失效的原因,由此提出了一种基于小世界邻域优化的局部线性嵌入(SLLE)算法.将复杂网络算法引入到流形学习中,利用小世界算法对LLE算法进行数据优化,并以最短路径和局部集群系数作为局部优化参数,解决了数据点不规则时以欧氏空间作为邻域判别标准在构建局部超平面造成嵌入结果扭曲的难题.通过3组标准测试数据集合比较了SLLE、LLE算法,结果表明SLLE算法的计算效果、鲁棒性、非理想数据的降维结果均优于LLE算法,且计算正确率至少提高10%.

Small World Neighborhood Optimized Local Linear Embedding Algorithm

By analyzing the invalidity reason of the local linear embedding(LLE) algorithm in case of the sparse data or the high noise data,small world neighborhood optimization LLE algorithm(SLLE) is proposed based on the complex networks theory.The data in LLE are optimized using the small world algorithm,and the shortest path and the local neighbor set clustering coefficients are used as the local parameters.As a result,the problem of the embedding distortion using only local linear patch of the manifold to define...