首页> 外国专利> HIGH ACCURACY CLASSIFICATION METHOD AND SYSTEM FOR ENORMOUS VOLUME OF DATA DESCRIBED WITH MULTIPLE VARIABLES

HIGH ACCURACY CLASSIFICATION METHOD AND SYSTEM FOR ENORMOUS VOLUME OF DATA DESCRIBED WITH MULTIPLE VARIABLES

机译:多变量描述的数据量庞大的高精度分类方法和系统

摘要

A method and a device for classifying information which can be expressed withmultivariable by using a computer with high precision at high speed inaccordance with similarity, a method for permitting a computer to practice aprocedure for classifying infotmation which can be expressed withmultivariables with high precision at high speed in accordance withsimilarity, a program for executing the method, and a computer-readablerecorded medium on which the program is recorded. For example, the method forclassifying data on input vectors with high precision by a computer bynonlinear mappimg includes the following steps ((a) - (f)) (a) Input vectordata is inputted to a computer, (b) an initial neuron vector is set, (c) theinput vectors are classified into respective neuron vectors, (d) the neuronvectors are updated into structures similar to the structures of the inputvectors classified into the respective neuron vectors and the structures ofthe input vectors classified into the neighborhoods of the neuron vectors, (e)the steps ((c), (d)) are repeated until a predetermined learning times isreached, and (f) the input vectors are classified into the neuron vectors andoutputted.
机译:用于对可以用以下方式表达的信息进行分类的方法和设备通过使用高速,高精度的计算机进行多变量运算根据相似性,一种允许计算机练习信息分类的过程,可以用高速,高精度的多变量相似性,用于执行该方法的程序以及计算机可读记录程序的记录介质。例如,用于通过计算机通过以下方法对输入向量上的数据进行高精度分类非线性mappimg包括以下步骤((a)-(f))(a)输入向量数据输入到计算机,(b)设置初始神经元矢量,(c)输入向量分为各自的神经元向量,(d)神经元向量被更新为类似于输入结构的结构向量分类为各自的神经元向量和结构输入向量分类为神经元向量的邻域,(e)重复步骤((c),(d))直到预定的学习时间为(f)将输入向量分类为神经元向量和输出。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号