International Journal of Special Education - Volume 41 - Issue 2 - 2026

Sep 2, 2026

Upload by Anda

A study presents an Intelligent Adaptive Learning and Monitoring System (IALMS) for learners with special educational needs (SEN). It defines SEN personas and assigns mean usefulness scores, then details module functions: a Learner Profile Engine for disability-aware segmentation, a Content Repository & Adaptation Engine using NLP to simplify and adjust reading levels with audio/visual alternatives, and a recommendation and pathway engine for ZPD-calibrated sequencing. Real-time learning analytics monitor engagement and risk to trigger early intervention workflows.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

IALMS Persona Profiles

Adaptive behaviours and usefulness scores

IALMS Module Architecture

Learner Profile Engine (LPE)

Content Repository & Adaptation Engine (CRAE)

ML Recommendation and Pathway Engine (MRPE)

Real-Time Learning Analytics Dashboard (RLAD)

Faculty & Admin Coordination Interface (FACI)

Disability Categories and Adaptations

Dyslexia

ADHD

Autism Spectrum Disorder

Visual impairment

Hearing impairment

Dyscalculia

| Persona | SEN Profile | Mean Usefulness Score (1–5) | Key IALMS Adaptive Behaviours Demonstrated | | --- | --- | --- | --- | | Aisha, 22, MBA Year 1 | Dyslexia + ADHD | 4.58 (SD = 0.19) | ALPS audio preference captured; case text simplified to GR Level 8; chunked pathway; microassessments; attention-lapse monitoring | | Rajan, 25, BBA Year 3 | ASD Level 1 | 4.52 (SD = 0.22) | Systemising style flag; explicit task structure; visual schedule overlay; sequential pathway with clear prerequisites; solo assessment alternative | | Priya, 28, Executive MBA | Partial Visual
Impairment | 4.64 (SD = 0.17) | Screen-reader optimised content; all graphics alt-texted; audio-first pathway; text-only assessments; full content accessibility confirmed | | IALMS Module | Primary ML Algorithm(s) | Input Data | SEN Accommodation Function | | --- | --- | --- | --- | | Learner Profile Engine (LPE) | K-means Clustering; Rule-based LNF | ALPS survey; preassessment; accessibility preferences | Disability-aware learner segmentation; needs flagging with explicit consent | | Content Repository & Adaptation Engine (CRAE) | NLP Pipeline (BERT, T5); Flesch-Kincaid scoring | Curriculum content; multimedia; assessment items | Text simplification; multimodal content generation; reading level adjustment; audio/visual alternatives | | ML Recommendation | Hybrid Collaborative | Interaction logs; | Personalised content sequencing; | | & Pathway Engine | Filtering; Deep Q- | completion rates; | ZPD-calibrated difficulty; | | (MRPE) | Network (DQN) | performance scores | engagement optimisation | |
Real-Time Learning |
LSTM Networks; | Time-series engagement | Academic risk prediction; | | Analytics Dashboard | Random Forest; | data; assessment results; | trajectory monitoring; overload | | (RLAD) | Isolation Forest | session logs | detection; early intervention alerts | | Faculty & Admin Coordination Interface (FACI) | Explainable AI (SHAP); dashboard visualisation | RLAD outputs; institutional data; faculty inputs | Interpretable risk alerts; accommodation workflow; curriculum adaptation recommendations | | Disability Category | LPE Profile Flag | CRAE Adaptation | MRPE Pathway Adjustment | RLAD Monitoring Focus | | --- | --- | --- | --- | --- | | Dyslexia | High extraneous load; low reading fluency | Text simplification; audio alternatives; font spacing | Reduced readingheavy sequences; extended time allocation | Reading speed deceleration; incomplete content access | | ADHD | Variable attention; high engagement variance | Shorter chunks; microassessments; gamification elements | Interleaved variety; spaced repetition scheduling | Session length; offtask navigation; assessment timing | | Autism Spectrum
Disorder | Systemising
preference; routine sensitivity | Explicit structure; visual schedules; reduced ambiguity | Sequential pathways; explicit prerequisite mapping | Rigid navigation; deviation from expected patterns | | Visual
Impairment | Screen reader dependency; audio preference | Full audio narration; alttext; no image-only content | Audio-first sequences; textbased assessments only | Screen reader event logs; content access success | | --- | --- | --- | --- | --- | | Hearing Impairment | Caption dependency; no audio-only access | Auto-captioned video; visual-first content | Captioned multimedia; textbased discussion options | Caption accuracy; video completion rates | | Dyscalculia | Quantitative processing difficulty | Numerical scaffolding; worked examples; calculator support | Extended quantitative pacing; additional worked LOs | Quantitative performance; timeon-task in finance modules |

FAQs

What is the role of the Learner Profile Engine (LPE) in IALMS?

LPE segments learners by disability-aware flags and accessibility preferences. It uses survey and preassessment inputs and obtains explicit consent for needs flagging.

How does the Content Repository & Adaptation Engine (CRAE) adapt learning content?

CRAE applies NLP techniques to simplify text and adjust reading levels, and generates multimodal alternatives. It supports audio-first access and ensures graphics have appropriate alt-text.

How does IALMS monitor learner risk during real-time learning?

RLAD analyzes time-series engagement and session logs to predict academic risk and detect overload or navigation issues. It triggers early intervention alerts and informs accommodation workflows through FACI.

Sep 2, 2026

Upload by Anda