JAD-IT-UZ: AN AI-POWERED PLATFORM FOR TEACHING ITALIAN AS A FOREIGN LANGUAGE WITH DOMAIN-SPECIFIC VOCABULARY TRACKS
https://doi.org/10.5281/zenodo.23414588
Kalit so‘zlar
artificial intelligence, Italian as a foreign language, CEFR assessment, trilingual terminology database, adaptive learning, speech recognition, e-learning platform, domain-specific vocabulary.Annotasiya
This paper presents the architecture and design rationale of JAD-IT-UZ, an AI-powered platform for teaching Italian as a foreign language, developed jointly by Uzbek and Italian university partners. The platform combines a four-layer client–content–AI–data architecture with a trilingual Uzbek–Italian–English terminology database and domain-specific vocabulary tracks in medicine, tourism, and business. We describe the passive and active AI components (text-to-speech, contextual translation, speech scoring, adaptive tutoring), the CEFR-aligned assessment design, and outline four open technical decisions — cloud versus on-premises inference, native versus cross-platform mobile development, the assessment method, and build-versus-buy for AI components — that will shape the project’s development roadmap.
Foydalanilgan adabiyotlar ro‘yhati
Council of Europe. (2020). Common European Framework of Reference for Languages: Learning, Teaching, Assessment – Companion Volume. Council of Europe Publishing.
Wainer, H. (Ed.). (2000). Computerized Adaptive Testing: A Primer (2nd ed.). Routledge.
Chapelle, C.A., & Sauro, S. (Eds.). (2017). The Handbook of Technology and Second Language Teaching and Learning. Wiley-Blackwell.
Nation, I.S.P. (2013). Learning Vocabulary in Another Language (2nd ed.). Cambridge University Press.
JAD-IT-UZ Project Consortium. (2025). Technical Task Breakdown and Architecture Specification [Internal project document]. Uzbekistan State University of World Languages & Università per Stranieri di Perugia.
