中国民航飞行学院学报

行业语料库构建是行业人工智能发展的基础,词频分析是行业语料库构建的关键步骤。本文以当前民航广泛使用的航空情报汇编资料(Aeronautical Information Publication,AIP)为语料提取源,提出了一种基于文本特征提取的航空情报语料词频分析方法。首先在对航空汇编资料原始结构分析基础上加入领域词典优化分词效果,随后对词频统计结果采用词频g指数的方法确定高频词阈值,对提取结果进行共词聚类分析。利用航空资料汇编样本进行方法验证,实验结果表明:本文所提出的方法能有效对航空汇编资料进行文本特征提取,为领域语料库应用奠定基础。

The construction of domain corpus is the basis for the development of industry artificial intelligence and word frequency analysis is a key step in the construction of domain corpus.This paper takes Aeronautical Information Publication(AIP) as the research object,which is widely used in civil aviation,proposes a word-frequency analysis method of aviation intelligence corpus based on text feature extraction.Firstly,on the basis of the analysis of structure of AIP,the domain dictionary is added to optimize the word segmentation effect.Then,the word frequency g-index method is used to determine the high-frequency word thresholds and the extracted results are subjected to word frequency statistics and co-word clustering analysis.The method is validated by using the compiled samples of AIP and the experimental results show that the proposed method can effectively extract textual features of aeronautical information publication and lay the foundation for constructing a corpus in the field of aeronautical information.