Digital Corpus Linguistics Based on Natural Language Processing (NLP) for English Islamic History Terminology in Indonesia

Abstract In recent years, there have been a growing number of English academic articles studying the history of Indonesian Islamic. However, the core terms used in these articles including proper nouns such as ulama, pesantren, tarekat, and sultanate have never been systematically sorted out and analyzed in combination with a corpus, leaving the actual usage of these words in academic circles unclear.To fill this gap, this study combines AntConc and Orange Data Mining to specifically analyze a pre-assembled collection of articles. This collection contains a total of 100 articles, all published between 2023 untill 2026, which have been integrated into a corpus available for analysis. After all articles were integrated, a total of ten distinct thematic directions were sorted out from this complete corpus. The top-ranked themes are those related to historiography and intellectual history, which have the largest number of related articles, ranking first among all themes. Instead of drawing conclusions solely relying on the results of a single tool, this study cross-validates the word frequency and collocation data output by AntConc with the clustering and topic modeling results produced by Orange. Finally, a transferable method template is compiled, which anyone can apply to combine computational methods and corpus-indexing-based research approaches to analyze professional terms in the field of religious history.