Citation Network (Citation Map): What It Is and How to Make One Free

Quick answer

A citation network is a graph of research papers where each dot (node) is a paper and each line (link) shows a connection: one paper cites another, two papers cite the same references (bibliographic coupling), or two papers are cited together (co-citation). Drawn on screen it is called a citation map. Make one free: search a topic on Ocean of Papers and click Build map.

What is a citation network?

Every research paper has a reference list. When paper A lists paper B in its references, A cites B. If you take many papers and draw every one of these citation links, you get a citation network: a web of papers (nodes) joined by citations (links).

The idea is old. In 1955 Eugene Garfield proposed a citation index for science, so that a reader could start from one paper and find every later paper that cited it (Garfield 1955, Science). That idea became the Science Citation Index and, later, every "Cited by" link you see in search engines today.

A citation map is simply a citation network drawn so you can look at it. People also call it a citation graph, a reference graph, a research paper graph or a literature map. Whatever the name, the job is the same: show the literature connections around a topic so you can see which papers are central, which belong together and which are new.

Why it helps:

• Find the key papers faster. Papers that many others cite stand out.
• Find papers keyword search missed. A paper can use different words but still be cited by, or cite, the same work.
• See the history of an idea. Older papers sit on one side, newer follow-ups on the other.
• Check your reading list. If your literature review leaves out a paper that half the network cites, you probably need it.

Updated October 2026.

Citation vs co-citation vs bibliographic coupling

Not every line in a citation map means "A cites B". Most tools use three classic types of link. A small example with four papers makes them easy to tell apart.

Annotated example: four papers, A to D

• Paper C (2010) is an early methods paper.
• Paper A (2015) cites C.
• Paper B (2016) also cites C. A and B do not cite each other.
• Paper D (2022) is a review. It cites both A and B.

Now look at the links this creates:

1. Direct citation: A → C, B → C, D → A, D → B. One paper lists the other in its references. This is the plain citation link.
2. Bibliographic coupling: A and B. A and B never cite each other, but both cite C, so they share a reference. The more references two papers share, the stronger the coupling. M. M. Kessler described this in 1963 (Kessler 1963, American Documentation).
3. Co-citation: A and B again. Paper D cites A and B together, so A and B are co-cited. The more later papers cite two works together, the stronger the co-citation. Henry Small described this in 1973 (Small 1973, JASIS); Irina Marshakova proposed the same idea independently in the same year.

So A and B are linked twice, for two different reasons: they look back at the same source (coupling) and later authors look at them together (co-citation). That is a strong sign they belong to the same research thread, even though neither cites the other. A keyword search might never show you both.

One practical difference: bibliographic coupling is fixed the day a paper is published, because its reference list does not change. Co-citation keeps growing as new papers cite the pair, so it works best for older papers and is weak for very new ones.

Citation graph vs citation tree vs citation map

These words get mixed up in search results and tool names. They overlap a lot, but there are small differences worth knowing.

A citation tree is the one that is most often misunderstood. Real citation data is rarely a clean tree, because one paper can be cited by many and can cite many. When people say "citation tree" they usually mean the ancestors of one paper (what it cites, and what those cite) or its descendants (who cites it, and who cites them). In a tool, that is the "references" or "cited by" view of one seed paper.

Literature connections is the plain-English version of all of this: how papers on a topic relate to each other.

How to read a citation map

A citation map can look like a bowl of noodles the first time. Read it in this order:

1. Size of the dot = how often the paper is cited. Big dots are well-known papers. In Ocean of Papers, bigger means more cited.
2. Colour = how new the paper is. In Ocean of Papers, darker means newer. Some other tools use a colour scale for year too, so check the legend.
3. Lines = a connection. A line can mean a direct citation, shared references, co-citation or a shared author. In Ocean of Papers you can click a line to see why the two papers are linked.
4. Clusters = sub-topics. Groups of dots packed close together usually share many references and form one research thread. Gaps between clusters often show different methods, populations or schools of thought.
5. Seminal papers = big, older, central. A large, light (older) dot that many lines reach is usually a foundational paper you should cite.
6. Newer follow-ups = small, dark, at the edge. Recent papers have had little time to collect citations, so they look small even when they matter. Do not ignore them; they show where the topic is going.

Two more habits help. First, switch to a timeline if your tool has one: putting publication year on one axis and citations on the other makes "old and important" versus "new and rising" obvious. Second, open the paper before you trust its position. A map shows connections, not quality. A heavily cited paper may be cited because others disagree with it.

How to make a citation map free (step by step)

You do not need special software or an account to make a citation network. Here is how to do it with Ocean of Papers, which is free for unlimited search and 5 new citation maps per month (mode switches and expansions do not count).

1. Search your topic, or paste a DOI. One search covers OpenAlex, PubMed, arXiv, bioRxiv, medRxiv and Europe PMC (250M+ papers). Pasting a DOI finds the exact paper.
2. Pick a strong seed paper. Choose one that is clearly on topic. A review or a well-cited paper gives a fuller map. You can also select up to 10 papers in the results to map them together.
3. Click Build map. You get about 30 papers by default; choose 15, 30 or 50.
4. Choose a mode. Similar shows the closest related work around the seed. Prior works shows what the seed cites, its "ancestors". Derivative works shows papers that cite it, its "descendants".
5. Switch Graph ↔ Timeline. The timeline plots publication year against citations (log scale), so seminal papers and new follow-ups separate clearly.
6. Click papers and lines. Click a line to see why two papers are linked. Use the Explore panel (Similar work, Earlier work, Later work, These authors) and Add to map, or Add as seed to grow the map around a second paper.
7. Use the side list. The Prior works and Derivative works tabs list the earlier and later papers for the whole map in one place.
8. Open PDFs and export. Open free, legal open-access PDFs where they exist, export the map as PNG, or export the papers as BibTeX, RIS or CSV for your reference manager. You can also copy a share link.

For a full walkthrough of every button, use the tool guide: free literature mapping tool.

1. Search or paste a DOI — 6 sources, 250M+ papers, no account needed
2. Pick a seed paper — Or select up to 10 papers to map together
3. Build map — 15, 30 or 50 papers
4. Choose a mode — Similar, Prior works or Derivative works
5. Try the Timeline — Year × citations shows seminal vs new papers
6. Click a line — See why two papers connect
7. Export — PNG, BibTeX, RIS or CSV, or share a link

Tips for a better citation map

• Start from the right seed. A narrow, recent paper gives a small map. A review article or a well-cited paper gives a richer one. If you have no seed yet, search your topic first and pick the most relevant result.
• Use Prior works to find the roots of a topic for your introduction, and Derivative works to find what is new for your discussion.
• Map two or three seeds. Selecting several papers together, or adding a second seed, fills gaps that one seed misses.
• Check the shared-author links. They help you spot one research group's papers, which is useful but can make a topic look bigger than it is.
• Save as you go. With a free account you can save papers into projects with notes and tags. Discover then suggests papers your project is missing, with reasons such as "cited by 4 of your papers".

Writing up what you found? Our guides on how to write a literature review and how to find papers that cite a paper take the next steps.

Other citation mapping tools

Ocean of Papers is not the only way to make citation maps. Other well-known options:

• Connected Papers: a similarity graph from one seed paper, built from co-citation and bibliographic coupling over Semantic Scholar data; 5 free graphs a month. See Ocean of Papers vs Connected Papers.
• ResearchRabbit: collections of seed papers with similar, earlier and later work.
• Litmaps: timeline-style maps plus Monitor alerts for new papers.
• Inciteful: free citation network analysis with rankings, no account.
• Open Knowledge Maps: topic clusters from a keyword (not a citation graph).
• VOSviewer and similar desktop programs: for large bibliometric studies (thousands of papers, co-authorship and keyword maps), usually from a database export.

Prices, free limits and data sources are compared side by side in best literature mapping tools and Connected Papers alternatives. For whole-field bibliometric maps, see science mapping tools.

Limitations of citation networks

Citation maps are useful, but they are not a judgement of quality. Keep these limits in mind:

• Citation lag. New papers have had no time to be cited. A brilliant 2026 paper will look tiny next to a 2010 paper, and co-citation links barely exist for it yet. Use bibliographic coupling (shared references) and the Derivative works view to find new work.
• Coverage gaps. Every map is only as good as its database. Books, theses, conference papers, non-English journals and older papers may be missing or have incomplete reference lists. Different tools use different data (Ocean of Papers maps mainly from OpenAlex; Connected Papers uses Semantic Scholar), so the same seed can give slightly different maps.
• Self-citation. Authors cite their own earlier work, often for good reasons. Heavy self-citation can make a group of papers look more connected or influential than the wider field thinks it is. Shared-author links help you notice this.
• Citations are not agreement. Papers are cited to be criticised, corrected or refuted. A retracted paper can still be heavily cited.
• Field differences. Medicine and biology cite far more than mathematics or the humanities, so "highly cited" means different things in different fields.
• Rich get richer. Well-known papers keep getting cited partly because they are well known. Check smaller nodes too.

Always read the actual papers, and combine the map with a normal database search. If you need the full text, see how to download research papers free.

FAQ

What is a citation network?

A citation network is a set of research papers connected by citations. Each paper is a node and each link shows that one paper cites another, or that two papers are related through shared references (bibliographic coupling) or by being cited together (co-citation).

What is citation mapping?

Citation mapping is building and reading a visual map of a citation network to find key papers, related work and new follow-ups. You start from one or more seed papers and the tool draws the papers around them.

Is a citation graph the same as a citation map?

Mostly, yes. "Citation graph" or "reference graph" is the network itself, often said in maths or computer-science terms. "Citation map" is the same network drawn on screen. Many tools use the words interchangeably.

What is a citation tree?

A citation tree usually means the papers one paper cites, and what those cite (its ancestors), or the papers that cite it, and what cites those (its descendants). In Ocean of Papers these are the Prior works and Derivative works modes.

What is the difference between co-citation and bibliographic coupling?

Bibliographic coupling links two papers that cite the same earlier work (they share references). Co-citation links two papers that are cited together by a later paper. Coupling is fixed at publication; co-citation grows over time.

How do I make a citation map for free?

Search your topic or paste a DOI on Ocean of Papers, choose a paper and click Build map. Search is unlimited and you get 5 new citation maps a month free, without an account. Inciteful and Connected Papers (5 graphs a month) also have free options.

How do I find literature connections between two papers?

Map both papers together (select them in search results, or use Add as seed), then click the lines between them to see whether they share references, are co-cited, cite each other or share authors.

Are citation maps reliable?

They are reliable for finding connected papers, but not for judging quality. New papers look small because of citation lag, coverage differs between databases, and self-citation or critical citations can inflate a paper's size.

Sources

• Garfield, E. (1955). Citation indexes for science: a new dimension in documentation through association of ideas. Science, 122(3159), 108–111. doi:10.1126/science.122.3159.108
• Kessler, M. M. (1963). Bibliographic coupling between scientific papers. American Documentation, 14(1), 10–25. doi:10.1002/asi.5090140103
• Small, H. (1973). Co-citation in the scientific literature: a new measure of the relationship between two documents. Journal of the American Society for Information Science, 24(4), 265–269. doi:10.1002/asi.4630240406

Ready to try it? Build a free citation map from any search result. Updated October 2026.

Related guides

• Free literature mapping tool (citation maps)
• How to find papers that cite a paper
• Connected Papers alternatives: free tools that connect papers
• Science mapping tools compared
• How to write a literature review (with example PDFs)

Search free research papers · Build a citation map from any search result · Browse all research guides