Conflict Interest is a smart system that helps detect academic or financial conflicts of interest among conference participants. It scans institutional affiliations, co-authorship history, and financial ties to flag potential conflicts automatically—streamlining a process that’s often manual and tedious.
🧑🤝🧑 Co-Authorship Detection – Identifies shared publications within the past 2 years using Google Scholar data.
🏢 Affiliation Matching – Flags current or past institutional overlaps using public and scraped data.
💸 Financial Conflict Detection – Connects individuals to potential financial ties where disclosure data is available.
📊 Conflict Scoring – Uses a weighted algorithm to assign a conflict score to each pair of individuals.
🖥️ User-Friendly Interface – View detected conflicts and their explanations in an intuitive UI.
The system operates through three primary phases:
- Scrapes or integrates data from:
- Google Scholar (publication + co-authorship)
- LinkedIn or institutional pages (affiliations)
- Public financial disclosures (if available)
- Extracts features like:
- Past or current shared institutions
- Co-authorship in the past 2 years
- Advisor–advisee relationships
- Financial ties from disclosures
- Computes a conflict score based on these features
- Input names of conference participants
- Display conflict results in a simple dashboard
- Explain reasoning behind each detected conflict
- Languages: Python, JavaScript
- Data/NLP: Pandas, Scikit-learn, Requests
- Frontend: HTML, CSS, JavaScript (Chrome Extension)
- APIs: Google Scholar (scraped or unofficial API), LinkedIn, public data sources
- Academic conference organizers
- Peer review managers
- Journal editors ensuring unbiased review
- Anyone vetting professional conflicts at scale
Built by Yujun Ge