Correlating Natural Language Parser Performance With Statistical Measures Of The Text
KI'09: Proceedings of the 32nd annual German conference on Advances in artificial intelligence(2009)
摘要
Natural language parsing, as one of the central tasks in natural language processing, is widely used in many AT fields. In this paper; we address art issue of parser performance evaluation, particularly its variation across datasets. We propose three simple statistical measures to characterize the datasets and also evaluate, their correlation to the parser performance. The results clearly show that different parsers have different performance variation and sensitivity against these measures. The method can be used to guide the choice of natural language parsers for new domain applications, as well as systematic combination for better parsing accuracy.
更多查看译文
关键词
different performance variation,natural language parsers,natural language parsing,natural language processing,parser performance,parser performance evaluation,better parsing accuracy,different parsers,AI field,central task,natural language parser performance,statistical measure
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络