半监督数据解码多候选结果准确提取仿真

Simulation of Accurate Extraction for Multiple Candidate Results Based on Semi-Supervised Learning

为有效对数据解码后的多候选结果准确提取,需要对数据解码后的多候选结果特征项进行分类提取。但当前方法对多候选结果提取时,无法得到准确的结果。为解决上述问题,提出一种基于半监督学习的数据解码多候选结果准确提取方法。方法采用非线性方法将半监督判定分析运用到数据特征提取当中,在利用已知类别和未知类别对数据解码多候选结果数据进行模型参数训练和学习,将学习后的数据解码多候选结果数据参数模型进行特征提取,根据得到的特征项将数据解码的多候选结果分类,最终得到准确的数据提取。实验结果证明,所提方法在对多候选结果数据提取时拥有较高速度和准确性。

In order to effectively extract the multiple candidate results after data decoding,it is necessary to classify and extract the feature items of multiple candidate result after data decoding.The current method cannot get the accurate results.Therefore,a method for accurately extracting multiple candidate results after the data decoding based on semi-supervised learning was proposed.This method used the nonlinear method to apply the semi-supervised judgment analysis to the data feature extraction.And then,the method used the known category and the unknown category to train and learn the model parameters of the data of multiple candidate result after the data decoding.Moreover,the features of parametric model of multiple candidate result of data decoding after the learning were extracted.According to the feature items,multi-candidate results were classified.Finally,the accurate data extraction was completed.Simulation results prove that the proposed method has higher speed and accuracy in extracting data of multiple candidate result.