公开(公告)号：US20160188694A1

公开(公告)日：2016-06-30

申请号：US14907610

申请日：2013-07-31

Applicant: HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.

Inventor： David Lehavi ,  Sagi Schein ,  Amir Globerson ,  Shai Shalev-Shwartz ,  Roi Livni

IPC: G06F17/30

CPC classification number: G06F16/285 ,  G06F16/2365 ,  G06F16/278 ,  G06K9/6217 ,  G06K9/6218 ,  G06K9/6269 ,  G06K9/6287 ,  G06K2209/01

Abstract: A method, system and storage device are generally directed to determining for each of a plurality of data points, a neighborhood of data points about each such data point. For each such neighborhood of data points, a projection set of polynomials is generated based on candidate polynomials. The projection set of polynomials evaluated on the neighborhood of data points is subtracted from the plurality of candidate polynomials evaluated on the neighborhood of data points to generate a subtraction matrix of evaluated resulting polynomials. The singular value decomposition of the subtraction matrix is then computed. The resulting polynomials are clustered into multiple clusters and then partitioned based on a threshold.

Abstract translation: 一般来说，方法，系统和存储设备针对多个数据点中的每一个确定关于每个这样的数据点的数据点的邻域。 对于每个这样的数据点邻域，基于候选多项式生成多项式的投影集。 在数据点附近评估的多个候选多项式中减去在数据点附近评估的多项式的投影集，以生成评估结果多项式的减法矩阵。 然后计算减法矩阵的奇异值分解。 所得到的多项式被聚集成多个簇，然后基于阈值进行分区。