CAJ | 학술논문

为了进一步提高Web信息抽取的准确性和效率，通过分析传统中文关键词抽取方法，文章提出了一种基于BP神经网络的中文关键词抽取方法。该方法在分析和提取术语特征的基础上，给出了确定网络隐层节点数的表达式和多个术语特征表达式，以此确定网络参数，实现中文关键词的抽取。实验结果表明，该方法查全率、查准率和查找性能较高，具备较好的应用前景。
위료진일보제고Web신식추취적준학성화효솔，통과분석전통중문관건사추취방법，문장제출료일충기우BP신경망락적중문관건사추취방법。해방법재분석화제취술어특정적기출상，급출료학정망락은층절점수적표체식화다개술어특정표체식，이차학정망락삼수，실현중문관건사적추취。실험결과표명，해방법사전솔、사준솔화사조성능교고，구비교호적응용전경。
In order to further enhance the accuracy and efficiency of Web information extraction ,a Chi-nese keyword extraction method based on back propagation(BP) neural network is proposed by analy-zing of the traditional Chinese keyword extraction method .The network hidden layer nodes expres-sions and terminology characteristic expression are given on the basis of analyzing and extracting term characteristic so as to determine the network parameters and achieve the Chinese keyword extraction . The experimental results show that the proposed method possesses high recall ratio ,precision ratio and look-up performance ,and it has a good application prospect .