摘要
为了更好的提高政务工作的生产效率,提出了编辑距离、改进的空间向量模型以及特征词库相结合的一种改进的文本相似度算法模型。改进算法充分考虑了政务文本的特征,以及特征词分类及权重影响,有效降低了计算的维度、缩小了计算的范围,大大提高了应用系统中检索数据返回的速率。系统运行结果也表明了改进的文本相似度算法在检测相似政务文本方面效果和性能有显著提高。
In order to improve government work efEciency of production, an improved text similarity algorithm model which was combined of edit distance and improved vector space model and feature words, was proposed in this paper. Full consideration of the characteristics of government text, as well as the characteristics of word classification and weight, improved algorithm can efficiently reduce the computational dimension, reduce the range of computing, and greatly increase the rate of retrieval of data returned in the application system. The effectiveness and accuracy of the improved algorithm are verified by the results of the system operation.
出处
《信息技术与信息化》
2016年第7期49-52,共4页
Information Technology and Informatization
基金
北京市科学技术情报研究所"萌芽"计划(所自立课题)--面向政务系统的文本相似度研究
