Scholarlink

<p>This study examines how AI-driven concordance analysis can identify syntactic patterns in aviation emergency communication, with the goal of informing ICAO-aligned training materials. Routine aviation exchanges follow standard phraseology, but emergencies demand flexible, context-specific language. Miscommunication in such situations can compromise safety. Corpus linguistics and speech act theory provide complementary frameworks for understanding how pilots and air traffic controllers use language in crises. A specialized 14,000-word corpus of 83 authentic transcripts of related emergencies was compiled. #LancsBox V6.0 retrieved concordance lines, and ChatGPT generated lexical bundles, which were compared with manual extractions using the Jaccard Similarity Index. Lexical bundles were analyzed for syntactic patterns and communicative functions using Austin’s and Searle’s speech act models. The AI and manually generated bundle lists showed 91.7% similarity, confirming high reliability of AI assisted corpus analysis. Key syntactic features included imperatives, modal verbs, elliptical forms, and technical noun phrases, serving functions such as alerting, requesting, and instructing. Speech act analysis highlighted their operational significance. AI assisted corpus analysis effectively uncovers authentic linguistic patterns in aviation emergencies, enabling the design of scenario-based, ICAO-aligned training. Future work should expand to other emergency types, integrate multimodal data, and pilot-test materials to validate pedagogical impact.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

From Corpus to Cockpit: Generating Emergency Lexical Bundles in Aviation Context

  • 期刊论文

  • 发表日期：

    2026年3月17日

查看全文
  • Deni Sapta Nugraha,
  • Eneng Uswatun Hasanah,
  • Ratna Dewanti,
  • Ifan Iskandar

摘要

This study examines how AI-driven concordance analysis can identify syntactic patterns in aviation emergency communication, with the goal of informing ICAO-aligned training materials. Routine aviation exchanges follow standard phraseology, but emergencies demand flexible, context-specific language. Miscommunication in such situations can compromise safety. Corpus linguistics and speech act theory provide complementary frameworks for understanding how pilots and air traffic controllers use language in crises. A specialized 14,000-word corpus of 83 authentic transcripts of related emergencies was compiled. #LancsBox V6.0 retrieved concordance lines, and ChatGPT generated lexical bundles, which were compared with manual extractions using the Jaccard Similarity Index. Lexical bundles were analyzed for syntactic patterns and communicative functions using Austin’s and Searle’s speech act models. The AI and manually generated bundle lists showed 91.7% similarity, confirming high reliability of AI assisted corpus analysis. Key syntactic features included imperatives, modal verbs, elliptical forms, and technical noun phrases, serving functions such as alerting, requesting, and instructing. Speech act analysis highlighted their operational significance. AI assisted corpus analysis effectively uncovers authentic linguistic patterns in aviation emergencies, enabling the design of scenario-based, ICAO-aligned training. Future work should expand to other emergency types, integrate multimodal data, and pilot-test materials to validate pedagogical impact.