Asosiy kontentga oʻtish
proceeding

Using AI to Bridge the Gap between Academic Assessments and Real-World Skill Application

Bektash BerdiyevInternational Islamic Academy of UzbekistanOrzigul NematovaNamangan State UniversityBehzod SoipovBakhtiyor AbsattorovTashkent branch of the University of Business and Science
2026en
Ochiq kirishOltin

Litsenziya: cc-by

ABI

Annotatsiya

The assessment-reality gap-the systematic divergence between what academic assessment systems measure and what real-world professional performance requires-is one of the most consequential and least resolved problems in educational quality assurance. This paper proposes an AI Framework for Bridging Assessment to Real-World Application (AFBARA) integrating a real-world task simulation and employer competency mapping engine, a reflective practice analytics and skill translation module, and a portfolio-based evidence and career readiness gateway, evaluated across 68 educational institutions and industry partners (universities$\mathbf{n} \boldsymbol{=} \mathbf{2 8}$, industry employer partners$\mathbf{n} \boldsymbol{=} \mathbf{2 4}$, professional certification bodies$\mathbf{n}=\mathbf{1 6})$processing$\mathbf{2, 6 8 4, 0 0 0}$assessment-to-employment records (1,878,800 training / 805,200 test, 70/30 stratified split) and 186,200 graduate-employer competency matchings across 6 industry sectors over a 40-month evaluation horizon. The AFBARA achieves a composite real-world skill bridging effectiveness score of$\mathbf{r}=\mathbf{0. 9 3 9}$against expert-validated employer-endorsed competency benchmarks, outperforming ML-only gap analysis platforms$(\mathbf{r}=\mathbf{0. 8 9 0}$; +5.5%$[(0.939-0.890) / 0.890])$, industry assessment tools$(\mathbf{r}=\mathbf{0. 6 3 8}$; +47.2% [(0.939-0.638)/0.638]), and traditional academic assessments$(\mathbf{r}=\mathbf{0. 4 7 8}; \quad+\mathbf{9 6. 4 \%} \quad[(\mathbf{0. 9 3 9 - 0. 4 7 8}) / \mathbf{0. 4 7 8}])$. Deployment outcomes: +62.2% real-world task performance prediction, +49.2% employer satisfaction score, -46.2% time-to-competency, +33.6% curriculum-industry alignment, +26.8% graduate employability index, and +56.6% portfolio-based competency demonstration. Sub-domain performance: Real-World Task Simulation Scoring$(\mathbf{r}=\mathbf{0. 9 6 7})$, Employer Competency Mapping (0.961), Reflective Practice Analytics (0.949), and Career Readiness Index (0.923). Domain-specific knowledge decay in highly specialised disciplines degrades AFBARA accuracy to 0.907 (-3.4% [(0.939-0.907)/0.939]), and temporal skill gap from assessment-to-employment delays exceeding 12 months to 0.902 (-3.9% [(0.939-0.902)/0.939]), with deployed mitigation protocols for both conditions.

Hali tarjima qilinmagan

Identifikatorlar

Iqtiboslar va manbalar