Logo Lanfrica
  • Accueil
  • Atlas
  • Analyses
  • Documentation
  • Sign in

© 2026 Lanfrica. Tous droits réservés. Tous les droits d'auteur des ressources affichées sur le site Web Lanfrica appartiennent aux détenteurs de droits d'auteur d'origine, sauf indication contraire explicite.

siltanukifilie/aims-ktt-ai-math-tutor

Domaine:

educationnatural language processing

Type de record:

software
Créateur:
sil
Hôte:
Offline AI Math Tutor for Early Learners (Ages 5–9) — Adaptive, Multilingual, Low-Resource EdTech Solution | AIMS KTT Hackathon | Siltanu Alemu # 🧠 AI Math Tutor for Early Learners (Offline, Multilingual, Adaptive) ## Overview This project presents an offline AI-powered math tutor designed for children aged 5–9 in low-resource environments. The system delivers foundational numeracy instruction through a combination of visual representations, audio prompts, and interactive feedback, while adapting dynamically to each learner’s progress. The tutor operates entirely offline, runs on CPU-only devices, and supports multilingual and code-switched input, making it suitable for real-world deployment in settings with limited connectivity and shared devices. --- ## Problem Context Across many low-resource settings, children complete early primary education without mastering basic numeracy skills. By ages 8–9, many learners still struggle with: - number recognition - simple arithmetic operations - problem-solving skills The core challenge is not access to schooling, but the lack of **personalized instruction**, compounded by language transitions (e.g., moving from Kinyarwanda to English). --- ## Solution The system delivers a complete end-to-end tutoring experience: - presents math problems using visuals and audio - accepts responses via text or speech (optional ASR) - detects and adapts to the child’s language - provides immediate feedback and guided correction - repeats questions when needed to reinforce learning - adapts difficulty using learner performance The focus is on **teaching**, not just evaluation. --- ## Core Capabilities ### 1. Adaptive Learning (Knowledge Tracing) A Bayesian Knowledge Tracing (BKT) model tracks mastery across skills: - counting - number sense - addition - subtraction - multiplication - division - word problems The system updates learner state after each interaction and selects the next item accordingly. --- ### 2. Multilingual & Code-Switched Interaction The tutor supports: …

Visit

github.com

Languages

Kinyarwanda

Similaires

siltanukifilie/aims-ktt-g5-s2t13-recommenderDrUkachi/ktt-math-tutor-datasiltanukifilie/aims-ktt-day2-g2-t2-2-grant-matcherFizzy9913/Offline-Math-Tutor-Aimansouro/aims-ktt-grid-outage-syntheticthieradams/senegal-cm1-math-tutor

siltanukifilie/aims-ktt-g5-s2t13-recommender

Made in Rwanda Content Recommender - AIMS KTT Hackathon 2026 # Made in Rwanda Content Recommender #

DrUkachi/ktt-math-tutor-data

Data artefacts for the AIMS KTT Hackathon Tier-3 submission S2.T3.1 AI Math Tutor for Early Learners

siltanukifilie/aims-ktt-day2-g2-t2-2-grant-matcher

Multilingual Grant & Tender Matcher for African entrepreneurs — ranks relevant grants/tenders by bus

Fizzy9913/Offline-Math-Tutor-Ai

AI-powered mathematics tutoring platform for Ghana JHS students (B7, B8, B9) with offline capabiliti

mansouro/aims-ktt-grid-outage-synthetic

This dataset is synthetically generated for the AIMS KTT Hackathon (Tier 2.3 - Grid Outage Forecaste

thieradams/senegal-cm1-math-tutor