Course 08

Capstone: Develop Your Model for Real-World Impact

The curriculum culminates in a comprehensive project that requires you to synthesize all previous learning. Build an end-to-end AI solution for real-world African challenges in healthcare, agriculture, climate, finance, or education.

Key Topics

  • Independent research and problem formulation
  • End-to-end model development pipeline
  • Real-world application focus
  • Responsible AI in African contexts
  • Project presentation and showcasing

Learning Outcomes

  • Transition from student to independent researcher
  • Apply technical skills to novel African problems
  • Demonstrate ability to build responsible AI solutions
  • Curate and represent data (Course 02)
  • Select architecture, train, fine-tune, align, and optimize

Capstone Project Areas

Low-Resource Language Translation

Develop models for African languages underrepresented in global datasets

Agricultural Intelligence

Predict crop yields or identify plant diseases based on local data

Healthcare Diagnostics

Fine-tune models for diseases prevalent in Sub-Saharan Africa

Climate Resilience

Analyze local climate data to predict weather patterns or flood risks

Financial Inclusion

Build credit scoring proxies for unbanked populations

Recommended Datasets

18 datasets are aligned with this course for hands-on exercises and projects.

Malaria Cell Images

Parasitized vs Uninfected Blood Cell Images (NIH)

image-classification computer-vision 27,558 images (~339MB)
English (Metadata)

KaraAgro AI Maize

Annotated Maize Disease Images from Ghana

object-detection image-classification 16,552 images (~1.5GB)
English (Metadata)

AfriMed-QA

Pan-African Medical Question Answering

medical-qa domain-adaptation 15,000+ QA pairs
English (African context)

Financial Inclusion in Africa

Bank Account Prediction Dataset

classification tabular ~33,000 rows
English (Metadata)

African Climate Change

Temperature Data for 5 African Countries

time-series forecasting Daily records (1980-2023)
Numeric

CryCeleb

Infant Cry Analysis from Nigeria

audio-classification healthcare 6.5 hours, 786 infants
Infant Cry (Audio)

GROW-Africa

African Crop Yield Data

regression yield-prediction 500,000+ observations
Numeric

AI4KSL

Kenyan Sign Language Dataset

sign-language-recognition computer-vision 20,000 videos
Kenyan Sign Language

Education in Africa

UNESCO African Education Statistics

data-analysis regression Tabular (Country-level)
Numeric

South Africa Crop Type

Fused Sentinel-1/2 Imagery for Crop Mapping

semantic-segmentation data-fusion ~10GB
English (Metadata)

Ghana Croplands Agronomy

Biophysical & Yield Measurements for Northern Ghana Maize

regression yield-prediction <100MB
English (Metadata)

African Medical Records (OCR)

Nigerian Handwritten Clinical Records for OCR

ocr computer-vision <1K records
English (Nigeria)

Nigerian Colposcopy Images

Cervical Cancer Screening Images from Nigeria

image-classification computer-aided-diagnosis 3,356 images
English (Metadata)

Med-Convo-Nig

Simulated Nigerian-Accented Medical Consultations

asr speech-recognition 4.2 hours audio
English (Nigerian accent)

Snapshot Serengeti

Camera-Trap Wildlife Images from Tanzania

image-classification object-detection ~5TB full (use a subset)
English (Metadata)

African Next Voices

Multilingual South African Speech (3000+ hrs)

asr speech-recognition 3,000+ hours
isiZulu isiXhosa Sesotho Setswana +3 more

Kencorpus

Kenyan Languages Corpus: Swahili, Dholuo, Luhya

machine-translation pos-tagging 5.6M words / 177 hours
Swahili Dholuo Luhya

Common Voice (Swahili & Hausa)

Crowdsourced Read Speech for African Languages

asr speech-recognition Swahili ~20GB; Hausa smaller
Swahili Hausa

About This Course

The capstone project is your opportunity to apply everything you’ve learned to solve a real problem. You’ll work on a project relevant to African contexts and present your work at a showcase event.

Project Themes

Choose a theme that addresses challenges in your country or region:

  • Healthcare: Disease prediction, health information systems
  • Agriculture: Crop yield prediction, pest detection
  • Finance: Fraud detection, financial inclusion
  • Education: Adaptive learning, content generation
  • Climate: Weather prediction, environmental monitoring
  • Governance: Public service optimization, transparency tools
  • Language Technology: Translation, NER, sentiment analysis

Capstone Requirements

  1. Problem Definition: Clear problem statement with measurable goals
  2. Dataset: Use curated datasets from this catalog or collect your own
  3. Solution: End-to-end implementation using course techniques
  4. Evaluation: Rigorous evaluation with appropriate metrics
  5. Responsible AI: Document ethical considerations and limitations
  6. Presentation: Showcase your work at the HEI event

Resources

Browse the datasets in this catalog for inspiration and data sources. Each dataset includes project ideas tailored for capstone projects.