Skip to content

User_manual.md

Latest commit

 

History

History
126 lines (87 loc) · 3.28 KB

User_manual.md

File metadata and controls

126 lines (87 loc) · 3.28 KB

Relatenta — User Manual

Overview

Relatenta visualizes research collaboration networks. Search for researchers via OpenAlex, ingest their publication data, and explore interactive network graphs and heatmaps.

Prerequisites

  • Python 3.10+
  • Modern web browser
  • Internet connection (for OpenAlex)

Installation

git clone https://github.com/Denny-Hwang/Relatenta.git
cd Relatenta
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate
pip install -r requirements.txt

Running

streamlit run streamlit_app.py

Opens at http://localhost:8501.

Workflow

1. Search for Authors

In the sidebar under "OpenAlex Search":

  1. Enter a researcher name (e.g., "Geoffrey Hinton")
  2. Click "Search"
  3. Review the results — each card shows:
    • Paper count, citation count
    • Institution and country
    • H-index
    • ORCID (if available)
    • Top 3 research topics with confidence scores

Disambiguation: Many researchers share the same name. Use institution, H-index, and research topics to identify the correct person before ingesting.

2. Ingest Data

  1. Select one or more authors from the search results
  2. Adjust "Max works per author" (default: 200)
  3. Click "Ingest Selected"
  4. Wait for ingestion to complete — the sidebar shows a summary of imported works

You can search and ingest multiple times to build up your dataset.

3. Visualize — Graph Tab

  1. Select a layer: authors, keywords, orgs, or nations
  2. Set year range and minimum edge weight
  3. Click "Build Graph"

Graph interaction:

  • Drag to pan the view
  • Click nodes to select
  • Scroll to zoom
  • Press SPACE to toggle physics simulation

Focus Mode (optional):

  • Use "Find Author IDs" / "Find Keyword IDs" expanders to look up IDs
  • Enter comma-separated IDs in the focus input
  • Choose "Full Network" (highlight) or "Focus Only" (isolate)

4. Visualize — Heatmaps Tab

  1. Select type: author_keyword or nation_nation
  2. Set year range
  3. Click "Compute Heatmap"
  4. Darker colors = stronger relationships

5. Export / Restore

  • Export: Click "Export CSV" in the sidebar to download a ZIP file
  • Restore: Upload a previously exported ZIP via "Restore from Export"
  • Clear: Click "Clear All" to reset the database

CSV Import

You can also import data from CSV files:

Type Required Columns
works title, doi, year, venue, type, language, keywords (semicolon-separated)
authors work_title, work_doi, author_name, position
affiliations work_title, work_doi, author_name, org_name, country_code
keywords work_title, work_doi, term

Import works first, then authors/affiliations/keywords (they link to works by DOI or title).

Tips

  • Start with 1-2 researchers and expand gradually
  • Use year filters and edge weight thresholds to reduce graph complexity
  • For large datasets (500+ papers), use Focus Only mode
  • Always export before closing the browser

Troubleshooting

Problem Solution
Graph too cluttered Increase edge weight minimum
No nodes in graph Lower edge weight or widen year range
Data lost after refresh Use Export/Restore to persist data
Slow ingestion Reduce max works per author

License

MIT License.