Research Paper Map (Find Related Papers on a Citation Graph)

About the Research Paper Map (Find Related Papers on a Citation Graph)

Give one to five papers you know — by DOI, arXiv id, PubMed id, OpenAlex id or title, or from a BibTeX or RIS file — and the map finds the papers most related to them. It collects each seed’s references, the 200 most cited papers that cite it and OpenAlex’s related works, scores every candidate by bibliographic coupling (the references two papers share) and co-citation (the papers that cite both), and draws the 40, 60 or 80 most related as a map: bigger circles are more cited, darker ones are newer, and lines join papers with much in common.

Beside the map, Prior works lists the earlier papers many of them build on, and Derivative works the later papers that cite many of them — the foundations and the newest work of the topic. An Authors view draws the co-author network. Open a paper to read its abstract, follow its DOI or open-access copy, or add it as a seed and draw the map again.

Common uses

• Starting a literature review from the one paper your supervisor gave you.
• Checking that a thesis or paper cites the important earlier work on its topic (the prior works list).
• Finding the newest work that builds on a classic paper (the derivative works list).
• Seeing which groups of authors work on a topic, and who works with whom.

How related papers are found

Two papers that cite the same earlier work are coupled (Kessler’s bibliographic coupling); two papers that are cited together by later work are co-cited (Small’s co-citation). Both are measured with Salton’s cosine: the number of shared references (or shared citing papers) divided by the square root of the product of the two papers’ totals, so a paper with a thousand references does not win by size alone. A candidate’s score is the mean of the two measures, averaged over the seeds.

Candidates are each seed’s references, the 200 most cited papers that cite it, and the works OpenAlex lists as related; the citing papers are also what co-citation is counted from. The map shows the highest scores, and joins two papers on it when their own similarity is high (each paper keeps at least its three strongest links). The layout is a force-directed drawing: linked papers pull together, all papers push apart, so clusters of closely related work sit together.

Where the data come from

Everything comes from OpenAlex, an open catalogue of scholarly works whose data are released under CC0: titles, authors, venues, years, citation counts, reference lists and abstracts. OpenAlex covers journals, conferences, books, theses and preprints across every field, but its reference lists are incomplete for some publishers, so a paper with few known references or citations gives a thin map: add another seed, or a better-known paper on the topic. OpenAlex allows each internet connection a number of requests a day; a map takes a handful, and the page says when your allowance is running low. MySmartCoPilot is not affiliated with or endorsed by OpenAlex; the map uses its public API.

Limitations

• Only papers in OpenAlex can be seeds, and the map is only as complete as OpenAlex’s reference lists and citation links.
• For a seed cited by thousands of papers, only the 200 most cited of them are collected, so very recent work may be under-represented.
• Up to five seeds and 80 papers on the map at a time.
• The similarity scores compare citation patterns, not the text: a related paper that cites different work can be missed.
• Collections are kept in this browser only: clearing the site’s data or using another browser leaves them behind.

Privacy

Sends data to online services

The DOIs, ids or titles you enter are looked up in OpenAlex (an open catalogue of scholarly works), straight from your browser and without cookies, together with the papers they reference and the papers that cite them. The ranking and the map are made on your device; collections stay in this browser.