CAJ | 학술논문

极限学习机ELM不同于传统的神经网络学习算法（如BP算法），是一种高效的单隐层前馈神经网络（SLFNs）学习算法。将极限学习机引入到中文网页分类任务中。对中文网页进行预处理，提取其特性信息，从而形成网页特征树，产生定长编码作为极限学习机的输入数据。实验结果表明该方法能够有效地分类网页。
겁한학습궤ELM불동우전통적신경망락학습산법（여BP산법），시일충고효적단은층전궤신경망락（SLFNs）학습산법。장겁한학습궤인입도중문망혈분류임무중。대중문망혈진행예처리，제취기특성신식，종이형성망혈특정수，산생정장편마작위겁한학습궤적수입수거。실험결과표명해방법능구유효지분류망혈。
ELM extreme learning machine is different from traditional neural network learning algorithm(such as BP algo-rithm), is a highly efficient Single hidden Layer Feedforward Neural network(SLFNs)learning algorithm. In this paper, ELM is introduced to Chinese web page classification task. Trait tree of web page is formed after pre-processing the Chi-nese web and extracting its characteristic information. Fixed-length coding is produced and took as input data of ELM. Experimental results show that the method can effectively classify web pages.