Science Mapping Tools (2026): VOSviewer, CiteSpace, Bibliometrix and More

Quick answer

Science mapping builds network maps of a research field from bibliographic data: co-citation, bibliographic coupling, co-word and co-authorship networks. The main free tools are VOSviewer (easiest, clear maps), CiteSpace (time-sliced, burst detection), Bibliometrix/biblioshiny in R (full bibliometric workflow), Sci2, CitNetExplorer and Gephi. For quick exploration from a search, web citation maps such as Ocean of Papers are faster.

What is science mapping?

Science mapping (also called the mapping of science or bibliometric mapping) uses bibliographic data to show how a research field is structured. In the words of a widely cited review, it aims "to build bibliometric maps that describe how specific disciplines, scientific domains, or research fields are conceptually, intellectually, and socially structured" (Cobo et al., 2011).

In practice, a science map is a network. The nodes are papers, authors, journals, institutions, countries or keywords. The links come from one of a few standard relations:

• Co-citation: two papers are linked when later papers cite them together. Frequently co-cited papers usually form the intellectual base of a field.
• Bibliographic coupling: two papers are linked when they cite the same references. Coupling works well for recent papers that have not been cited yet, so it is good for mapping research fronts.
• Direct citation: one paper cites the other. Used for citation networks and historiographs (how a field developed over time).
• Co-word (co-occurrence): two keywords or terms are linked when they appear in the same papers. This gives a conceptual or topic map.
• Co-authorship: two authors, institutions or countries are linked when they publish together. This gives a social or collaboration map.

Science mapping complements performance analysis (counting publications and citations). Performance analysis tells you how much; science mapping tells you how things are connected. For a gentler introduction to citation links, read what is a citation network.

Updated October 2026.

The science mapping workflow

Cobo, López-Herrera, Herrera-Viedma and Herrera reviewed the main science mapping software in Science mapping software tools: Review, analysis, and cooperative study among tools (Journal of the American Society for Information Science and Technology, 62(7), 1382–1402, 2011; doi:10.1002/asi.21525; author PDF). They describe a general workflow that still fits most projects today:

1. Data retrieval: export records from a bibliographic database (Web of Science, Scopus, Dimensions, OpenAlex, PubMed, Lens and others).
2. Preprocessing: remove duplicates, fix author and keyword variants, choose a time span, and reduce the data to the items that matter.
3. Network extraction: build the network (co-citation, coupling, co-word, co-authorship) from the cleaned data.
4. Normalization: adjust raw link counts with a similarity measure (for example association strength, cosine or Jaccard) so that big items do not dominate.
5. Mapping: lay out the network and find clusters.
6. Analysis: study the clusters, their themes, and how they change over time.
7. Visualization: produce the map, overlay, timeline or density view.
8. Interpretation: an expert reads the map and draws conclusions.

The review also notes that no single tool does everything well, which is why many researchers combine tools, for example Bibliometrix for cleaning and statistics and VOSviewer for the final map.

Science mapping tools compared

The table covers the established bibliometric mapping software and four web citation-map tools that students often use for quick exploration. Bibliometric software works on exported data files and gives you full control over the analysis. Web tools work from a search or seed paper and give you a map in seconds, with much less control. Facts for the web tools come from our literature mapping tools comparison (checked October 11, 2026).

VOSviewer

VOSviewer is developed by Nees Jan van Eck and Ludo Waltman at the Centre for Science and Technology Studies (CWTS), Leiden University, and is free to download. It was introduced in van Eck & Waltman (2010), Software survey: VOSviewer, a computer program for bibliometric mapping, Scientometrics 84(2), 523–538 (doi:10.1007/s11192-009-0146-3; free full text in PMC).

• Networks: publications, journals, researchers, organizations, countries, keywords or terms, linked by co-authorship, co-occurrence, citation, bibliographic coupling or co-citation.
• Data: files from Web of Science, Scopus, Dimensions, Lens and PubMed, reference manager files, and downloads through APIs including OpenAlex.
• Views: network, overlay (for example colouring by average year) and density.
• Strengths: the cleanest maps of all tools here, good with large networks, easy for beginners, and VOSviewer Online lets you share interactive maps.
• Limits: limited data cleaning (you use a thesaurus file to merge variants) and no statistics beyond the map.

CiteSpace

CiteSpace was created by Chaomei Chen (Drexel University) and is described in Chen (2006), CiteSpace II: Detecting and visualizing emerging trends and transient patterns in scientific literature, JASIST 57(3), 359–377 (doi:10.1002/asi.20317). Downloads and the current "Advanced" builds are distributed through the author's site, citespace.podia.com; check the current version and terms there.

• Strengths: time-sliced networks that show how a field changes, citation burst detection for sudden spikes of attention, cluster labelling, timeline and time-zone views, and measures such as betweenness centrality to find "turning point" papers.
• Data: built around the Web of Science format; other sources are converted inside the tool.
• Limits: a steeper learning curve and many parameters. Results depend heavily on settings (time slices, node selection), so report them clearly in your methods.

VOSviewer vs CiteSpace: choose VOSviewer when you want a clear, static map of structure (clusters of co-cited work, keyword themes, collaboration) and an easy start. Choose CiteSpace when your question is about change over time: emerging topics, bursts and the papers that connect periods. Many published bibliometric studies use both.

Bibliometrix and biblioshiny (R)

Bibliometrix is an open-source R package by Massimo Aria and Corrado Cuccurullo, described in bibliometrix: An R-tool for comprehensive science mapping analysis (Journal of Informetrics 11(4), 959–975, 2017; doi:10.1016/j.joi.2017.08.007). It is available on CRAN. Biblioshiny is its point-and-click web interface, which runs in your browser on your own computer, so you do not need to write R code.

• Data: imports from Scopus, Web of Science, Dimensions, OpenAlex, Lens, PubMed and Cochrane.
• Strengths: covers the whole workflow in one place: merging and cleaning data, descriptive statistics, performance analysis, co-citation, coupling, collaboration and co-word networks, thematic maps and thematic evolution. Because it is R, analyses can be scripted and reproduced.
• Limits: network pictures are less polished than VOSviewer's; many users export networks to VOSviewer or Gephi for the final figure.

Sci2, CitNetExplorer and Gephi

Sci2 (Science of Science Tool) is a free, modular toolset from the Cyberinfrastructure for Network Science Center at Indiana University for temporal, geospatial, topical and network analysis of scholarly data. Its website is marked as deprecated and points to GitHub for the latest version, and it no longer runs natively on current macOS (a Docker option is offered). It is still useful for teaching and for its wide range of algorithms, but it is not actively updated.

CitNetExplorer comes from the same CWTS team as VOSviewer and is free for non-commercial research and teaching. It focuses on direct citation networks of individual publications, laid out over time, which helps you trace how a field developed and delineate the literature on a topic. It works with Web of Science data. It was introduced in van Eck & Waltman (2014), Journal of Informetrics 8(4), 802–823 (arXiv preprint).

Gephi is free, open-source software for exploring and visualising any kind of network. It is not a bibliometric tool by itself: you build the network in Bibliometrix, Sci2 or your own scripts and then use Gephi for layouts, filtering, statistics and publication-quality styling.

Web citation map tools: Connected Papers, Litmaps, ResearchRabbit, Ocean of Papers

Web tools are not full science mapping software. They do not let you import a database export, choose a normalization or report reproducible parameters. What they do well is fast exploration: you start from a search or a paper and see its neighbourhood in seconds.

• Connected Papers builds a similarity graph around one seed paper using co-citation and bibliographic coupling over Semantic Scholar data (about). Free for 5 graphs a month.
• Litmaps draws timeline citation maps and is the strongest option for new-paper alerts (Monitor).
• ResearchRabbit grows collections from seed papers and shows similar, earlier and later work; it has a free plan and an optional paid tier.
• Ocean of Papers starts from a search across OpenAlex, PubMed, arXiv, bioRxiv, medRxiv and Europe PMC. From any result you build a map of about 30 papers (15, 30 or 50) in Similar, Prior works or Derivative works mode, as a graph or a timeline (year × citations). Links use the same relations as bibliometric software, direct citation, bibliographic coupling (shared references), co-citation and shared authors, and you can click a link to see why two papers connect. You can select up to 10 seed papers, then export the papers as BibTeX, RIS or CSV or the map as PNG.

Detailed comparisons: Connected Papers, Litmaps, ResearchRabbit, and alternatives in Connected Papers alternatives.

Bibliometric software or a quick citation map?

Use bibliometric software (VOSviewer, CiteSpace, Bibliometrix) when:

• You are writing a bibliometric analysis or science mapping study that will be peer reviewed
• You need hundreds or thousands of records from a defined database query, with a documented search string
• You must report methods: data source, time span, unit of analysis, counting method, thresholds and normalization
• You want co-authorship, country or keyword maps, not only paper-to-paper links

Use a quick citation map tool when:

• You are starting a literature review and want to find the key papers and clusters fast
• You have one or a few seed papers and want similar, earlier or later work (how to find papers that cite a paper)
• You do not have access to Web of Science or Scopus exports
• You want to check, before a formal study, which sub-areas exist and which keywords to use in your database query

A practical combination: explore a topic in Ocean of Papers to learn its structure and collect key papers, export them as RIS or BibTeX into your reference manager, and use what you learned to write a better search string for Web of Science, Scopus or OpenAlex. Then run the formal analysis in Bibliometrix or VOSviewer on that export. For open data, OpenAlex lets anyone export result lists as CSV from its website (up to 100,000 works per export, according to its help pages) and offers a free full-data snapshot.

Your first science map in VOSviewer (overview)

The exact labels change between versions, so follow the official VOSviewer manual for your version. The general steps are:

1. Export your data. Run your search in Web of Science or Scopus (or another supported source) and export the records with full records and cited references, in a format VOSviewer reads.
2. Start the wizard. In VOSviewer, click Create and choose to create a map based on bibliographic data.
3. Choose the source. Read data from bibliographic database files (or download through an API) and select your file.
4. Choose the analysis. Pick the type of analysis (for example co-citation) and the unit of analysis (for example cited references, sources or authors). The type sets the links; the unit sets the nodes.
5. Choose counting and thresholds. Select full or fractional counting and a minimum number of occurrences or citations so the map stays readable.
6. Check the items. Review the selected items and remove or merge ones that do not belong (a thesaurus file helps with spelling variants).
7. Finish and explore. Switch between network, overlay and density views, adjust clustering resolution, and save the map and its settings so your figure is reproducible.

Report the database, query, date of export, time span, analysis type, unit, counting method and thresholds in your methods section.

1. Export records — With cited references
2. Create → bibliographic data — Start the wizard
3. Select file / API — WoS, Scopus, etc.
4. Analysis + unit — e.g. co-citation of references
5. Thresholds — Keep the map readable
6. Explore views — Network, overlay, density

Good practice for science mapping studies

• Define the field first. A map is only as good as the search that built it. Pilot your search string and check that key papers are included.
• Clean before you map. Merge author name variants, singular/plural keywords and synonyms. Small cleaning steps change clusters a lot.
• Pick the relation that fits your question. Co-citation for the intellectual base, bibliographic coupling for current research fronts, co-word for themes, co-authorship for collaboration.
• Be careful with databases. Web of Science, Scopus, Dimensions and OpenAlex cover different journals and years. Do not mix exports without de-duplicating, and say which source you used.
• Report parameters. Thresholds, normalization and clustering resolution all change the picture. Readers should be able to reproduce your map.
• Interpret with expertise. Cluster labels from algorithms are a starting point; read the core papers in each cluster before naming it.
• Get the papers legally. For papers you need to read, see websites to download research papers for free.

FAQ

What is science mapping?

Science mapping is the use of bibliographic data to build network maps that show how a research field is structured: its intellectual base (co-citation), research fronts (bibliographic coupling), themes (co-word) and collaboration (co-authorship). It is a core method in bibliometrics and scientometrics.

What is the mapping of science used for?

Researchers use maps of science to describe a field for a review or bibliometric study, find influential papers and authors, spot emerging topics, and study collaboration between institutions and countries. Funders and libraries use them for research evaluation and strategy.

What are the best science mapping tools?

The most used free tools are VOSviewer, CiteSpace and Bibliometrix (with biblioshiny). Sci2, CitNetExplorer and Gephi are useful for specific tasks. For quick exploration from a search, web tools like Ocean of Papers, Connected Papers, Litmaps and ResearchRabbit are faster but less configurable.

VOSviewer vs CiteSpace: which should I use?

Use VOSviewer for clear maps of a field's structure and an easier start. Use CiteSpace when you need to show change over time, citation bursts and emerging trends. Many studies use both, often with Bibliometrix for data cleaning and statistics.

Is there free bibliometric mapping software?

Yes. VOSviewer is free, Bibliometrix and Gephi are free and open source, Sci2 is free, and CitNetExplorer is free for non-commercial research and teaching. CiteSpace is free to download, with Advanced builds distributed through the author's site.

Can I do science mapping without Web of Science or Scopus?

Yes. VOSviewer and Bibliometrix both work with OpenAlex, a free and open database, and with sources such as Dimensions, Lens and PubMed. OpenAlex results can be exported as CSV from its website or accessed through its free API.

Is a citation map the same as a science map?

They are related. A citation map from a web tool shows papers around a seed or a search, using links such as direct citation, co-citation and bibliographic coupling. A science map usually covers a whole field from a large, defined dataset, and can also map authors, journals, countries or keywords with documented parameters.

Summary

For a formal study, pick VOSviewer for clear structure maps, CiteSpace for change over time, and Bibliometrix for a reproducible end-to-end workflow; use Gephi when you need custom figures. For fast exploration before or alongside that work, a web citation map such as Ocean of Papers shows you the key papers and clusters from a single search, and exports them as BibTeX, RIS or CSV to take further. Learn more in our literature mapping tool guide.

Updated October 2026.

Related guides

• Citation network: what it is and how to make one free
• Best literature mapping tools compared
• Connected Papers alternatives: free tools that connect papers
• 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