摘要
阐述了词条匹配系统的设计思路、系统的总体架构及实现的技术方法,在现有理论基础上,借助盘古分词技术对特征来源文本进行分词与去噪,通过计算关键词的特征权重使得特征词条选取最优;运用先进的特征匹配算法对系统进行了详细设计和实践开发.对实验数据的分析结果表明,该系统对图书辅助分类校验具有一定的实际应用价值.
The design idea of the system, the overall architecture of the system, the realization of the technical methods were elaborated. On the basis of existing theory, Pangu word segmentation technology was used for text segmentation and denoising. Then the calculation of feature weight keywords was used to determine feature term optimal selection. The advanced feature matching algorithm of the system were employed to provide detailed design and practice development. The analysis showed that the system has a certain practical value for the book classification and verification.
出处
《南通大学学报(自然科学版)》
CAS
2017年第3期15-19,共5页
Journal of Nantong University(Natural Science Edition)
基金
江苏省现代教育技术研究立项课题(2015-R-42479)
关键词
词条
词条匹配
特征权重
文本特征
vocabulary entry
vocabulary entry matching
feature weight
text feature