节点文献
受限领域中文文本主题标引系统研究
Study of Chinese Textual Topic Identification System in Domain-dependent
【摘要】 文中介绍了受限领域中文文本主题标引系统的设计原理。该系统基于领域知识库进行主题特征识别 ,根据知识库中概念间的关系进行特征项权重的计算。实现了基于词汇分析技术和基于领域知识计算相结合的方式下 ,受限领域中文文本内容主题概念的识别。实验结果显示 ,主题识别平均正确率为 79%。
【Abstract】 This paper presents the design principle of Chinese textual topic identification in domain-dependent. With domain knowledge applied to identify topic feature,and relevancy among concepts in domain knowledge utilized to calculate feature weight,Chinese textual topic is identified. The identification is based on the combination of lexical analysis and domain knowledge analysis. Experiment results show that the average accuracy of topic identification reaches 79 % by this approach.
- 【文献出处】 计算机应用 ,Computer Applications , 编辑部邮箱 ,2004年01期
- 【分类号】TP391.1
- 【被引频次】3
- 【下载频次】109