EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Adaptive AI Tutors Scale Personalized STEM Education Across Diverse Learner Populations

Zhenyu Luo, Haotian Ren, Matthew Carter

Computer Life · 02.04.2026 · 14 (1) · 27-33

DOI: 10.54097/wrdscr15 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer Life, online erschienen am 02.04.2026. Verfasst von Zhenyu Luo und 2 weiteren Autor:innen. Die Arbeit nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The persistent challenge of delivering individualized instruction across heterogeneous learner populations in science, technology, engineering, and mathematics (STEM) contexts has driven significant interest in artificial intelligence (AI)-powered adaptive tutoring systems. This paper examines how adaptive AI tutors can be effectively scaled to meet the cognitive, motivational, and demographic diversity of students engaged in STEM learning. Drawing on a mixed-methods research design that integrates machine learning-based knowledge tracing, reinforcement learning (RL)-driven content sequencing, and learner profile modeling, this study evaluates an adaptive intelligent tutoring system (ITS) deployed across three secondary and tertiary STEM courses involving a combined cohort of 847 students. Quantitative findings indicate statistically significant improvements in learning gains, task completion rates, and knowledge retention among students who engaged with the adaptive system compared to those in conventional instructional settings. Qualitative data further reveals that learners from under-resourced backgrounds and high-achieving students both reported increased perceived relevance and self-regulatory engagement when the system dynamically adjusted content difficulty and scaffolding level. The study also identifies equity-related concerns regarding data sparsity for underrepresented learner groups and proposes mitigation strategies anchored in fairness-aware machine learning. These findings contribute to an emerging understanding of how scalable adaptive AI tutoring can bridge achievement gaps in STEM education without sacrificing instructional depth or pedagogical coherence.

Zitieren

APA

Luo, Z., Ren, H., & Carter, M. (2026). Adaptive AI Tutors Scale Personalized STEM Education Across Diverse Learner Populations. Computer Life, 14(1), 27-33. https://doi.org/10.54097/wrdscr15

Vancouver

Luo Z, Ren H, Carter M. Adaptive AI Tutors Scale Personalized STEM Education Across Diverse Learner Populations. Computer Life. 2026;14(1):27-33. doi:10.54097/wrdscr15

BibTeX
@article{Luo2026,
  author = {Luo, Zhenyu and Ren, Haotian and Carter, Matthew},
  title = {Adaptive AI Tutors Scale Personalized STEM Education Across Diverse Learner Populations},
  journal = {Computer Life},
  year = {2026},
  volume = {14},
  number = {1},
  pages = {27--33},
  doi = {10.54097/wrdscr15}
}
Kostenlos starten →