An interactive visualization platform for exploring the Islam West Africa Collection β a dataset of 19,000+ documents on Islam and Muslims in West Africa.
π Access the Dashboard | π Dataset on Hugging Face
The IWAC Dashboard transforms a large scholarly dataset into an accessible, interactive exploration tool. It enables researchers to discover patterns, relationships, and trends across thousands of documents β without writing code or querying databases.
Designed for: Scholars in Islamic studies, African studies, religious studies, media studies, digital humanities, and related fields.
Languages: Fully bilingual interface (English/French) with real-time switching.
The entry point provides key statistics at a glance: total documents, language distribution, country coverage, and recent additions to the collection.
| Page | What You Can Discover |
|---|---|
| Country Distribution | Which West African countries have the most documentation? Interactive treemap showing document density by nation. |
| World Map | Geographic spread of the collection visualized on an interactive choropleth map. |
| Sources Map | Where are the newspapers and publication sources located? |
| Entity Geographic Footprint | Select any person, organization, or topic and see where they appear across the region. Track an imam's influence, an organization's reach, or a concept's geographic spread. |
| Page | What You Can Discover |
|---|---|
| Timeline | How has coverage of Islam in West Africa evolved over time? View growth trajectories, monthly additions, and identify periods of intensive documentation. |
| Categories Over Time | How has the composition of document types changed across decades? |
| References by Year | Publication patterns and bibliographic trends across time periods. |
| Page | What You Can Discover |
|---|---|
| Word Cloud & Frequency | What terms appear most frequently? Filter by country or year to see regional and temporal variations in terminology. |
| Word Co-occurrence | Which terms appear together? Identify semantic clusters and conceptual relationships in the literature. |
| Topic Modeling | What themes emerge from automated analysis? Browse detected topics and their prevalence. |
| Sensitive Terms | Track concerning or problematic terminology over time. Useful for critical discourse analysis and understanding media framing. |
| Page | What You Can Discover |
|---|---|
| Entity Network | How are people, organizations, places, and topics connected? Interactive graph showing relationships based on co-mentions in documents. |
| Spatial Network | Which locations are mentioned together? Geographic clusters and regional connections visualized on a map. |
| Page | What You Can Discover |
|---|---|
| Entity Index | Searchable directory of all persons, organizations, events, locations, and topics extracted from the collection. |
| Language Distribution | What languages are represented? Breakdown by document type and country. |
| Top Authors | Who has contributed most to the scholarly literature? Publication counts and activity periods. |
The dashboard helps researchers investigate questions such as:
Geographic patterns
Temporal dynamics
Actors and networks
Discourse analysis
Entity tracking
All data comes from the Islam West Africa Collection on Hugging Face, a curated dataset documenting how francophone West African newspapers have covered Islam and Muslims from the colonial period to the present.
The collection includes:
If you use this dashboard in your research, please cite:
Madore, FrΓ©dΓ©rick. Islam West Africa Collection Dashboard.
https://fmadore.github.io/iwac-dashboard/
# Install dependencies
npm install
# Start development server
npm run dev
# Build for production
npm run build
cd scripts
pip install -r requirements.txt
python generate_overview_stats.py # Run individual generators
src/routes/ β Page componentssrc/lib/components/visualizations/ β Chart and map componentssrc/lib/stores/ β State managementscripts/ β Python data generationstatic/data/ β Pre-computed JSON filesSee CLAUDE.md for detailed development guidelines.
This project is open source. The underlying dataset is available under its own license on Hugging Face.