1. Corpus construction for Persian language

The faGLOBE Persian Balanced Corpus is the first non lingua franca corpus in the global corpus cluster of Beijing Foreign Studies University, focusing on the collection of written modern Persian language materials. Persian is the official language of Iran and Tajikistan, as well as one of the main languages of Afghanistan. It belongs to the Indo European language family and uses a 32 letter pinyin script. The existing Persian language corpus is widely used in linguistic research, teaching, and artificial intelligence development. For example, Peycare synchronic corpus, specialized corpora such as news agency title corpus, learner corpus, syntactic corpus, oral corpus, and Persian handwritten discrete letter database.

2. Creation of faGLOBE Corpus

The faGLOBE corpus version 1.0 mainly collects original Persian language versions from Iran, excluding translated texts. The corpus has expanded the collection time range, with news corpora mostly consisting of the latest reports from 2022, while other genres of texts were first published in the past decade. In the process of building the corpus, the diversity of sources and themes of the corpus was taken into account, including multiple Iranian local media and academic corpora from different disciplines. When entering text, the length of the collected text in the corpus is about 2000 words, while the short text is composed of multiple similar corpora. All corpus is manually proofread and stored as a txt text file in UTF-8 encoding format. The faGLOBE corpus contains 500 texts with a length of 2000 words, stored as 798 language material sub files. The file names include language codes, genre codes, numerical codes, and letter codes. The corpus also records metadata annotations in Excel spreadsheet format, including 10 basic information such as file number, text type, number of texts, word count, author, publication year, publisher or newspaper name, corpus collector, article or book title, and website source.

3 Research Prospects

The faGLOBE corpus (version 1.0) provides Persian language learners with a wealth of learning resources, including 15 categories of texts such as news reports, editorials, religion, biography, technology, and official documents, which help expand language knowledge and enhance self-directed learning abilities. Teachers can use the word frequency function for vocabulary teaching, syllabus vocabulary selection, and auxiliary teaching. At the same time, as a teaching tool, it can display a large amount of corpus and examples to help students master the semantics and usage rules of vocabulary. Corpus is of great value to linguistic research. Researchers can analyze the distribution of Persian vocabulary, language style trends, word clusters, etc. through word frequency and collocation functions, and understand the common forms and rules of vocabulary collocation. In the future, the faGLOBE corpus will undergo deep processing such as morphological restoration, part of speech coding, syntactic annotation, and semantic annotation, providing different versions of the corpus to better identify the meaning of words in context.

conclusion

The faGLOBE corpus is the first Persian language corpus aimed at studying Persian through real language data and promoting a deeper understanding of Iran from multiple perspectives. Emphasizing resource co construction and sharing, technological innovation, enriching content and usage methods to support Persian language teaching and Iranian research.