针对多语种警务语料库在构建过程中遇到的资源分散和人工整理难的突出问题,提出了一种基于Web数据获取的多语种警务语料库的构建方法,该方法采用了频繁序列模式和文本分析技术实现了对网页中多语种语料库的自动抽取。经实验证明,该方法可进行多语言数据的自动采集,在少量人工参与的前提下,实现了多语言数据的大规模采集和处理。 Aiming at the prominent problems of multi-language police corpus, which are difficult to decentralize resources and manual collation, the paper proposes a construction method of multilingual police corpus based on web data acquisition. The method uses frequent sequence mode and text Analysis technology to achieve automatic extraction of multilingual corpus web pages. The experiment proves that the method can automatically collect multi-language data and realize the large-scale acquisition and processing of multi-language data with a small amount of manual participation.