Getting Started

To get started with VOSviewer, you may want to take a look at some of the sources of information listed below.

Video

Share this page

VOSviewer manual

The VOSviewer manual provides detailed and up-to-date information on the functionality of VOSviewer. The manual can also be accessed using the Manual button in VOSviewer.

Open VOSviewer manual

 

VOSviewer book chapter

VOSviewer is discussed in detail in a recent book chapter. The chapter provides a tutorial in which the use of VOSviewer is demonstrated in a step-by-step manner. The chapter is based on VOSviewer version 1.5.5. Some of the improvements and extensions introduced in the most recent versions of VOSviewer are therefore not covered in the chapter. In addition to VOSviewer, the chapter also discusses the CitNetExplorer tool for visualizing and analyzing citation networks of publications.

Open VOSviewer book chapter

VOSviewer paper

VOSviewer was originally introduced in a paper published in Scientometrics in 2010. In the meantime, VOSviewer has been improved and extended in many ways, so the paper is not fully up-to-date anymore. However, it still provides a useful introduction into the main ideas underlying VOSviewer.

Open VOSviewer paper

AIDA bookletAIDA booklet

The AIDA booklet introduces diverse case studies on research positioning and trend identification. These case studies are relevant for PhD candidates, researchers, group leaders, and policy makers. Many of the case studies make use of VOSviewer.

Open AIDA booklet

Comments

Mandatory fields
  • stephanie November 12th, 2019 4:07 pm
    Hi
    I have a question related to citing journals: if I am studying a specific journal's body of literature, how is it possible to find the other journals that are citing 'my' journal? I know on web of science for example how to get to that information but I was curious to see if I can also get it through VOSviewer, for example to get a map. Thx in advance
    Reply
  • Aynul November 8th, 2019 3:34 am
    Under "Types of analysis" i dont get active for "Bibliometric coupling", "citation" and "co-citation" options. Only co-occurrence and co-authorship are active. So i cannot perform analysis. Whats the remedy please?
    Reply
    • Ludo Waltman November 8th, 2019 7:43 am
      What kind of data are you working with? In the case of Web of Science and Scopus data, you need to make sure that references are included in the data that you download from these data sources. There are also data sources that don't provide references at all (e.g., PubMed and reference managers). When the latter data sources are used, it is not possible to create citation-based maps. See also the VOSviewer manual for more details.
      Reply
  • Anuja November 7th, 2019 6:44 pm
    Dear Mr. Waltman,
    I have .csv files exported from scopus and I want to conduct co-citation analysis. Although the options for analysis is inactive. Is there any way I can analyse co-citation network through .csv files.
    Thank You
    Anuja Joseph
    Reply
    • Ludo Waltman November 7th, 2019 7:11 pm
      You need to make sure to include references when exporting data from Scopus. This will enable you to perform co-citation analysis.
      Reply
  • Dario Sablone November 5th, 2019 12:11 am
    Dear Mr. Waltman,
    I am making a co-citation analysis of a csv file from Scopus. I have also created a Thesaurus file. The problem is that the software does not read the Thesaurus file correctly: it changes the labels to just some of the articles and not to all those that should be replaced by a correct one. If you know any reason why this happens I would be happy to hear it.
    Thank you very much,
    Dario Sablone
    Reply
    • Ludo Waltman November 6th, 2019 7:39 am
      Please send your files to vosviewer@cwts.leidenuniv.nl. I will then try to help you with this.
      Reply
  • Adel Mohebbi November 1st, 2019 6:12 am
    Dear Ludo Waltman,
    I have a CVS file from scopus and a plain text from Web Of Science . How can I use both of them in VOSviewer ?
    Or, what Kind of export format do u propose to use web of science and scopus data base(in one file for example) together in vosviewer.
    Thank you very much
    Reply
    • Suzi Dias November 1st, 2019 1:41 pm
      Dear Adel,
      I have the same question.
      When you get it, can you share with me, please?
      my e-mail is dias.suziferreira@gmail.com.
      Thank you very much.
      Reply
      • Ludo Waltman November 2nd, 2019 10:02 am
        Unfortunately, there is no easy way to combine data from Web of Science and Scopus. Combining the two data formats is highly challenging. In general, my recommendation is to work with data from a single data source.
        Reply
  • Bob Adler October 28th, 2019 7:12 pm
    Question from a newb...
    How does one download 7,800+ search citations from PubMed, without selecting them one at a time or exporting and then reassembling multiple clipboards with what appears to be 500 citation clipboard limit or the 200 citation display limit? Arrgh!
    Have to be certain this question has been asked before and someone has answer / method
    Thank you very much
    BOB
    Reply
    • Ludo Waltman October 28th, 2019 9:59 pm
      I am afraid I don't understand your question. It is easy to export large numbers of records from PubMed. You need to export the records in a MEDLINE file, as explained in the VOSviewer manual (p. 27-28). I don't know what you mean by clipboard and display limits.
      Reply
      • Bob October 28th, 2019 10:11 pm
        Thank you for your response. Yes, figured out large downloads.
        Reply
    • Adel Mohebbi November 1st, 2019 1:53 pm
      OK.sure
      Reply
  • Ammar Sani October 23rd, 2019 4:57 pm
    Dear Ludo Waltman,
    I am new to this software. I've read your manual and all previous comments for looking to answer my confusing. But, I do not find any of them yet.
    I am now creating a co-occurrence map. There is an item named 'adaptive learning' which is connected only with my main keyword, 'learning analytics'. Once I hover on the item, it states cluster 1, links 40, total link strength 59 and occurrences 16.
    May I know why the links counted as 40 even though there is only one link, which is the one mentioned above?
    Thanks
    Reply
    • Ludo Waltman October 23rd, 2019 5:59 pm
      I think you need to increase the number of lines that are shown in the visualization. This can be done in the options panel in the right area of the VOSviewer window.
      Reply
      • Ammar Sani October 23rd, 2019 6:13 pm
        Thanks for this response. Yes that the solution. Now they are all showing perfectly! Thanks once again.
        Reply
  • Majid October 20th, 2019 12:00 am
    I recently downloaded the software from your website (OS=windows 10), and for some reasons, the information panel is missing. possible it is hidden? any chance I can bring it back?
    Reply
    • Ludo Waltman October 20th, 2019 8:23 am
      The information panel is available only for specific types of maps, in particular for citation-based maps of documents. In the case of a map of documents, the information panel is used to provide information about a selected document (e.g., title, authors, and publication year). For all types of maps, the status bar is available to provide general information about a map and about specific items in a map.
      Reply
  • Kahlil Lawless October 18th, 2019 9:17 pm
    I'm trying to load up a Dimensions data-set with >53,000 rows to do a co-authorship bibliometic analysis. The file is ~300MB. I keep getting OOM errors. Any estimate how much computing power I would actually need to perform this?
    Reply
    • Ludo Waltman October 18th, 2019 11:39 pm
      I am not sure how much memory you would need for this. However, you may try to remove redundant columns from your Dimensions CSV file. Most columns are not needed for a co-authorship analysis. Reducing the file size might make it easier for VOSviewer to process the file.
      Reply
  • Osaze Omoregbe October 18th, 2019 3:38 pm
    Hi guys,
    After using VOSViewer software for sometimes, I made a video to help beginners understand the use of the software. You can watch the video via the link: https://www.youtube.com/watch?v=EIKdq8mBllo
    Reply
    • Ludo Waltman October 18th, 2019 11:36 pm
      Thanks!
      Reply
  • Muhammad Ajmal Khan October 17th, 2019 8:01 am
    i tried to upload file containing around 8 thousands records with with references downloaded from Scopus but facing problem of "Out of memory" consult manual , that I consulted where is written run jar file there is no Jar file, I am using vosview1.6.13 for windows
    Muhammad Ajmal Khan
    Head, Library Quality and Academic Accreditation
    Liaison Librarian, College of Bussiness Administration and College of Public Health
    Deanship of Library Affairs
    Building A3 Central Library|New Campus
    Imam Abdulrahman Bin Faisal University, Dammam, KSA
    M: 00966 (0) 540783455
    T: 00966 (0) 13 33 32809
    Reply
    • Ludo Waltman October 17th, 2019 8:46 am
      The JAR file can be downloaded from https://www.vosviewer.com/download using the link 'Download VOSviewer 1.6.13 for other systems'.
      Reply
  • Juan Pablo Usuga October 16th, 2019 11:24 am
    Hello Mr. Waltman,
    Thank you for your last answer,
    I have another question:
    How does VOSviewer deal with keywords with missing year information to estimate the "average pub. year" in the Overlay visualization? Is VOSviewer ignoring these entries?
    Thank you,
    J.
    Reply
    • Ludo Waltman October 19th, 2019 10:24 pm
      What is the data source that you are using? And are you creating a keyword co-occurrence map or a term co-occurrence map (i.e., a co-occurrence map based on terms extracted from titles and abstracts)?
      Reply
      • Juan Pablo Usuga Cadavid October 21st, 2019 11:42 am
        Hello,
        I am creating a keyword co-occurrence map from the "Author keywords" (so no term extraction from abstracts).
        Thank you,
        J.
        Reply
        • Ludo Waltman October 21st, 2019 1:45 pm
          And what is the data source you are using?
          Reply
          • Juan Pablo Usuga Cadavid October 21st, 2019 2:07 pm
            Hello,
            I'm using a CSV file from SCOPUS (where I actually put also some references from ScienceDirect that I had to transform from a RIS file, as ScienceDirect is not directly supported by VOSViewer).
            Thanks,
            J.
            Reply
            • Ludo Waltman October 23rd, 2019 6:55 am
              We checked the VOSviewer source code. It turns out that in the case of a keyword co-occurrence map a missing publication year is treated as a publication year that is equal to -1. This is a bug. We will fix this in the next update of VOSviewer. Thank you for bringing this issue to our attention.
              By the way, when working with Scopus data, there normally should be no missing publication years in the data.
              Reply
  • stephanie October 11th, 2019 9:44 am
    Hi all, I have a couple of questions, one related to co-occurrences and the other to the country information.
    1- Is it correct to say that the co-occurrence map is 'only' using the keywords information ? It would be nice to create a co-occurrence map using title and abstract data as well, is this something I can do by creating a corpus myself? Has anyone tried it? Does that work 'well' or are there some issues?
    2- Is the 'country' information, for example in the bibliographic coupling feature, related to the organization country (and not the author nationality)? For example an article authored by an author working at Berkeley University would be assigned to the USA, even if the author is not american ?
    thx in advance
    Reply
    • Ludo Waltman October 11th, 2019 9:47 am
      1. No, this is not correct. VOSviewer supports two types of co-occurrence maps, one based on keywords and the other based on terms extracted from titles and abstracts of publications.
      2. Yes, this is correct.
      Reply
      • stephanie October 11th, 2019 9:53 am
        So in VOSViewer the first option for co-occurrence, 'All keywords', includes titles + abstracts + author keywords?
        thx again
        Reply
        • Ludo Waltman October 11th, 2019 4:53 pm
          No, this option is just for keywords, not for terms from titles and abstracts. My suggestion is to take a look at Section 3.5.4 in the VOSviewer manual. This should clarify things.
          Reply
  • Juan Pablo USUGA CADAVID October 7th, 2019 4:19 pm
    I have two questions concerning the relatedness and one concerning the clusters:
    1) Could you please explain me the difference between the lines and the spatial position of two words in the graph? In other words, do the lines and the spatial position (closeness) tell the same information? (strong line = close elements and weak line = far elements)
    2) Do you have any paper where I can find more details about the estimation of this relatedness with the lines. If so, could you please give me the reference to read it?
    3) Do you have any paper where I can find more details about the creation of clusters? If so, could you please give me the reference to read it?
    Thank you in advance,
    J.
    Reply
    • Ludo Waltman October 11th, 2019 9:13 am
      1) Correct. The lines and the closeness provide approximately the same information. The proximity of two items approximately reflects the strength of the link between the two items.
      2) We don't have a paper specifically on this topic.
      3) See https://doi.org/10.1016/j.joi.2010.07.002 or https://arxiv.org/abs/1006.1032.
      Reply
  • Lotus Ramana September 30th, 2019 4:21 pm
    I am using VOSviewer for text mining, and I’m having difficulty understanding how the Occurrences attribute is calculated. According to the VOSviewer manual, “In the case of full counting, the Occurrences attribute indicates the total number of occurrences of a term in all documents.” This seems simple enough, but when I create a visualization based on text data, using the full counting method, I get a map which shows a particular term with “Occurrences: 4,” and I am certain that in my corpus the term appears 15 times. Can you suggest any reasons for this discrepancy? Many thanks, in advance.
    Reply
    • Ludo Waltman October 5th, 2019 10:49 pm
      It is difficult to answer your question without seeing your data. If you send your data to vosviewer@cwts.leidenuniv.nl and if you explain which term you are referring to, I can take a closer look and I will try to answer your question.
      Reply
      • Lotus Ramana October 14th, 2019 1:49 pm
        Thanks so much for your response and for clearing this up for me!
        Reply
  • Aamir Memon September 27th, 2019 3:06 pm
    I am struggling to define the terms "link" and "link strength" in my paper. In the network visulation map for author keywords, we had 66 keywords with 479 links and a total link strength of 1338 (minimum link strength = 6, maximum link strength = 456). How do we define it in simple terms?
    Thanks in advance!
    Best,
    Aamir
    Reply
    • Ludo Waltman September 28th, 2019 1:35 pm
      Please see my response below to the question by David Sanchez. This question is similar to yours. Chapter 2 in the VOSviewer manual also explains this issue. Let me know if you need additional information.
      Reply
      • Aamir Memon September 29th, 2019 6:19 pm
        Dear Dr. Waltman,
        I had already seen the response to the question by David Sanchez and had read the description in chapter 2 as well. I would, however, be grateful to get your response to this particular situation that I discussed in my question because the description is still unclear to me.
        Thank you in advance!
        Best,
        Aamir
        Reply
        • Ludo Waltman September 29th, 2019 7:28 pm
          There are 479 pairs of keywords between which there is a co-occurrence link. Each link has a link strength, which is the number of documents in which two keywords co-occur. The sum of the link strengths equals 1338.
          Reply
          • Aamir Memon September 29th, 2019 9:31 pm
            This means that in 66 keywords included, there were 479 pairs of keywords between which there was a co-occurrence link. For example, physical activity and sleep co-occur together making a link of co-occurrence. Is this a correct explanation?
            Each link has a link strength, which is the number of documents in which two keywords co-occur. The group of keywords co-occurred in a way that that the sum of the link strengths was 1338. Is that right?
            Thanks for the explanation!
            Best,
            Aamir
            Reply
            • Ludo Waltman October 5th, 2019 10:55 pm
              Yes, correct!
              Reply
  • Deepa September 26th, 2019 1:27 pm
    Dear Dr. Ludo Waltman,
    Is there any way that I can import my edited Scopus CSV excel file into VOSViewer? Currently, it is not allowing me to choose among the types of analysis i.e., by default the software is only running the co-authorship analysis.
    Thanks.
    Reply
    • Ludo Waltman September 28th, 2019 1:25 pm
      Our advice is not to edit Scopus CSV files in MS Excel. Although in principle it is possible to edit Scopus CSV files, in practice this almost always causes all kinds of problems.
      Reply
      • Deepa September 28th, 2019 7:00 pm
        Okay.
        But sir, if we require to clean the Scopus data first (to yield accurate results), can we convert the file into an excel sheet, make the desired changes and reconvert the file into csv format, before importing into VOSViewer?
        Reply
        • Ludo Waltman September 28th, 2019 8:26 pm
          I am afraid this will cause problems. The difficulty is that a cell in Excel has a maximum size. Certain fields in a Scopus CSV file, such as the reference field, sometimes exceed the maximum size of a cell in Excel. This is causing problems.
          Reply
          • DEEPA September 29th, 2019 10:54 am
            Alright, sir. I'll search for better alternatives then.
            Thank you so much for your valuable suggestions.
            Reply
  • Varsha September 26th, 2019 1:13 am
    I cant able to open the file and the message shows
    "while reading VOSviewer thersaurus file (line:1): PTT is not a valid coloumn header)".
    Kindly help me
    Thanks ,
    Varsha
    Reply
    • Ludo Waltman September 28th, 2019 1:23 pm
      The only valid column headers for the first line of a thesaurus file are ‘label’ and ‘replace by’. See Section 4.3 in the VOSviewer manual for more details.
      Reply
  • Kumar September 22nd, 2019 7:15 pm
    Dear Ludo Waltman,
    Is it possible to know the source documents of keywords such that the relevant papers can be separated for literature review and analysis?
    Reply
    • Ludo Waltman September 22nd, 2019 10:30 pm
      There currently is no easy way to obtain this information in VOSviewer. However, there is an option that you could use. When creating a keyword co-occurrence map, in the final step of the 'Create Map' wizard, a table is shown listing the keywords to be included in the map. By right-clicking on this table, it is possible to export the document-keyword relations to a file.
      Reply
      • Kumar September 23rd, 2019 3:10 pm
        Dear Ludo Waltman,
        Thank you for your immediate response. I could use the method and get needful information. I have another query. I have downloaded search results from both Web of Science and Scopus. Is there a method to develop a single keyword map from both these results. Secondly, whether the duplicate items in both the results affect the output. Is there a way to eliminate duplicates.
        Reply
        • Ludo Waltman September 23rd, 2019 3:58 pm
          I am afraid there is no easy way to merge data obtained from different data sources.
          Reply
          • Leo Soares September 24th, 2019 3:13 pm
            Hi there. Ive seen a paper where He merged both collections into mendeley to eliminate duplicates. But Im not sure if it was to prepare the data for VOSViewer or another software.
            Reply
  • Lurdes September 16th, 2019 11:26 pm
    Dr. Ludo Waltman,
    It's possible to know the Degree, Betweenness Centrality and Closeness Centrality in the co-citation network (Cited references).
    Thanks
    Reply
    • Ludo Waltman September 17th, 2019 12:04 am
      VOSviewer does not provide centrality statistics. I personally believe they are often misinterpreted and used incorrectly. In VOSviewer, you can save your map in a GML file or in a Pajek network file. You can then perform centrality calculations in Gephi or Pajek.
      Reply
  • xin September 7th, 2019 1:13 am
    Dear Dr. Waltman,
    Just wondering is there any way to use zotero exported document and put them into VOSviewer?
    Otherwise, do you know any approach that I can export zotero data into scoupus or web of science bibliography format? I could not find a way so far, but I really do not want to mannully assmble 150 citations again in scoupus to visualize it...
    Thanks a lot!
    Reply
    • Ludo Waltman September 8th, 2019 11:25 am
      Do you have DOIs for your documents? In that case, my advice would be to export the DOIs in a plain text file, to provide this file as input to VOSviewer, and to then use the Crossref or COCI API to obtain bibliographic data for your documents. An alternative approach could be to try to export your data from Zotero in a RIS file. This RIS file can then be provided as input to VOSviewer. Exporting your data in a file in the Web of Science or Scopus file format won't be possible.
      Reply
      • xin September 8th, 2019 8:37 pm
        Hi Dr.Waltman,
        I tried to export the RIS data from Zotero and imported it to VOSviewer. It really helps!!!
        Thanks!
        Best,
        Xin
        Reply
  • Ibtissam Zaza September 6th, 2019 11:50 pm
    Dear Dr. Waltman,
    I am new to VOSviewer. I am trying to create a map based on bibliographic data. I chose that option -> read data from reference manager files -> select files "EndNote". My question, if my library is in EndNote X9, how can I download a file with the extension .enw from EndNote? If you can direct me, I highly appreciate it!
    Reply
    • Ludo Waltman September 8th, 2019 11:28 am
      I am afraid I cannot help with this. I am not an EndNote user myself, so I cannot provide any guidance on how to use EndNote.
      Reply
  • Ibtissam Zaza September 6th, 2019 12:02 am
    Dear Dr. Waltman,
    I want to conduct co-word analysis over the library that I have on my EndNote. how can I link VOSviewer to EndNote?
    Reply
    • Niloofar Solhjoo September 16th, 2019 10:05 pm
      Hi
      First, you need to export all your references from Endnote. Save it as a Text File type and your output style must be "EndNote Export". (To enable this style in your EndNote software go to: Edit-->Output styles-->Open style manager--> tick the EndNote Export and close the window).
      Then open your text file in a Notepad and save as it again. this time your file extension must be like "file-name.enw" and save it as All Files type.
      There you go, now you have a file with .enw extension which is supported by VOSviewer.
      Reply
  • Natalia K August 22nd, 2019 10:43 am
    Dear Dr. Waltman, I've got the following question. While constructing co-authorship map, is there any way to integrate the link strength directly from the data? For example, the data looks like this: A and B co-authored X articles, B and C - Y articles, A and C - Z articles. Can VOS viewer process this kind of data to construct the map? Thank you in advance!
    Reply
    • Ludo Waltman August 23rd, 2019 8:00 am
      I am not sure if I fully understand your question. VOSviewer is indeed able to process this kind of data. This is the main purpose for which VOSviewer can be used.
      Reply
      • NK August 23rd, 2019 10:11 am
        Thank you for your reply, and please allow me to clarify. The number of co-authored articles is already in the file - it has been calculated, so that the VOSviewer doesn't have to do it like it normally does with the files from the Scopus and other databases. My trouble is with creating the .csv file that VOSViewer could recognize and figuring out which header I should use for the column with the number of co-authored articles. For now, VOSViewer counts each of those co-authored articles as one, rather than occurring X, Y, Z times.
        Reply
        • Ludo Waltman August 23rd, 2019 12:52 pm
          It seems you are trying to create a VOSviewer network file. For more information about these files, I recommend taking a look at the VOSviewer manual. Basically, a VOSviewer network file consists of three columns. In your case, the first two columns contain the names (or identifiers) of authors, while the third column contains the number of co-authored publications for a pair of authors. The columns don't have a header.
          Reply
          • NK August 28th, 2019 10:46 am
            Thank you! That's exactly what I was looking for. Have a good day!
            Reply
  • CONCETTA IOVINE August 8th, 2019 11:18 pm
    Hi, we are elaborate some network with your programma. We was no able to put the name in the network in capital letter. Is it possible? Thank you
    Reply
    • Ludo Waltman August 9th, 2019 8:25 am
      Please see the question by Eric R below. I think your question is similar to his.
      Reply
  • Jude 'Feranmi July 29th, 2019 6:42 pm
    Dear Waltman,
    Please I will like to know if Vosviewer can help with constructing the raw data from the WOS database that I downloaded on country collaborations using the address.
    I have about 50,000 articles I am trying to analyse and can't build the data by looking at the addreses and collaborations one after another. Is there a way Vosviewer can do this?
    Thanks in advance
    Reply
    • Ludo Waltman July 29th, 2019 8:41 pm
      What do you mean by 'constructing the raw data'?
      Reply
      • Jude 'Feranmi July 30th, 2019 12:13 pm
        I will like to run a regression analysis using a collaboration index. But for me to calculate the index, I need to know how many instances collaboration by a country with another country occured in a particular year.
        i.e a table that has a column for each country and the number of articles that the country partnered with. e.g Germany has 5 articles with Austria, 6 with Netherlands, 7 with Spain, 8 with France and so on
        I assume this data is what is used to construct the map and visualisations. Is it possible for me to get the data?
        Thank you sir.
        Reply
        • Ludo Waltman July 30th, 2019 2:08 pm
          Yes, this is possible. You need to create a co-authorship map of countries. You then need to save the map (using the 'Save' button on the 'File' tab) in a VOSviewer map file and a VOSviewer network file. The data that you are looking for can be extracted from these two files.
          Reply
  • David Sanchez July 25th, 2019 4:18 pm
    Hello,
    I struggle with the interpretation of links and link strength. E.g., when I do a citation analysis of authors, I get e.g. 700 citations, 750 links and link strength of 1200 of author X. The number of citations is derived from the database (e.g. Scopus). How are the links and total link strength derived exactly? What do the link and total link strength exactly stand for? Is it possible that an author has more links than citations?
    Thanks a lot!
    With best, David
    Reply
    • Ludo Waltman July 26th, 2019 8:20 am
      In your example, the focal author has citation links with 750 other authors. The focal author may be cited by these authors, may cite to these authors, or both. With some authors, the focal author has more than one citation link. This is reflected by the total link strength of 1200. In total, the focal author has 1200 citation links with the 750 authors with which it has citation links.
      Reply
      • David Sanchez July 26th, 2019 10:51 am
        Thank you so much for the explanation. I searched a lot, but I did not find such a good and simple explanation anywhere.
        Reply
  • ASHISH GOSAIN July 23rd, 2019 11:26 am
    Hi,
    I have a dataset on Patents granted for Bt COTTON at the USPTO FOR 2005-TO 2018. Ihave recorded the forward citations of the subs et of these Patents. Some of these are cited by Examiner and others are volunatry. Can this be visualised using VOS Viewer using an Excel File of the Data. Is there any coding required. Please consider a beginner's level in answering this question.
    Thanks,
    Ashish
    Reply
    • Ludo Waltman July 23rd, 2019 11:09 pm
      I am afraid there is no easy way to do this. You need to create your own map and network files, following the file formats documented in the VOSviewer manual. This can be fairly complex. It requires considerable data processing skills.
      Reply
  • Eric R July 18th, 2019 12:17 am
    For a bibliographic network visualization, is there any way to edit the author names so that they can appear in upper case or initial caps and/or without the first/middle initials? Thanks!
    Reply
    • Ludo Waltman July 19th, 2019 8:18 pm
      Yes, this can be done, not within VOSviewer but outside. You need to save your map in a VOSviewer map file. This is a plain text file with tab-delimited columns. The file can be opened in a spreadsheet program such as MS Excel. You can then manually edit the labels of the items in your map (i.e., the author names). When you save the file and then open it again in VOSviewer, the labels will be displayed in the desired way.
      Reply
  • Alihosein July 16th, 2019 10:35 am
    I am using vosviewer now, I want to add the thesaurus I have made to do co-ocurrence but the error appear. the error is below
    error while reading vosviewer thesarus file (line 26): incorrect number of columns
    I want to know what is the error and how can I fixed it?
    thank you
    Reply
    • Ludo Waltman July 16th, 2019 11:30 pm
      Please send your thesaurus file to vosviewer@cwts.leidenuniv.nl and I will try to offer some advice.
      Reply
      • Alihosein July 17th, 2019 6:11 am
        thank you so much for your attention
        I sent the thesaurus file i have made to vosviewer@cwts.leidenuniv.nl
        have a great day
        Reply
  • GG July 12th, 2019 6:28 pm
    Dear Ludo,
    Thanks for your comments and work.
    I performed a co-citation analysis (from Web of Science). I get the references (papers) with the cluster number. I export the result (save VOSViewer map file), visible with excel. Each reference has an ID.
    My question is: how can I generate a file that indicates ID, authors, journals... and abstract? (my goal is to perform a lexical analysis of abstracts while keeping the cluster number).
    Thanks again !
    Reply
    • Ludo Waltman July 13th, 2019 7:07 am
      I am afraid that abstracts are not available for the references in data downloaded from Web of Science. They are available only for the documents in data downloaded from Web of Science, not for the references in these documents. Perhaps you could do a bibliographic coupling analysis instead of a co-citation analysis, since a bibliographic coupling analysis is based on the documents in your data, not on the references in these documents.
      Reply
  • A.A. July 11th, 2019 10:37 am
    Dear all,
    How can I expand my map as in adding other items to my existing map?
    Thank you.
    Reply
    • Ludo Waltman July 12th, 2019 7:22 am
      VOSviewer does not have any functionality for expanding maps. However, you can do this manually by saving your map in a VOSviewer map file and by editing this file in a spreadsheet program such as MS Excel. In Excel, you can then manually add new items to your map.
      Reply
  • Lurdes July 7th, 2019 3:03 am
    Hello Ludo,
    It's possible, to know the articles with highest co-citation frequencies?
    Thank you,
    Lurdes
    Reply
    • Ludo Waltman July 7th, 2019 4:01 pm
      When you have created a co-citation map of documents, the total link strength of a document indicates the total number of co-citations the document has with other documents included in your map.
      Reply
    • Lurdes July 8th, 2019 12:41 am
      Thank you.
      Another subject, the authors' country calculation on VOSviewer (Citation/Countries) is different from that calculated by the Web of Science, for some countries.
      Reply
  • Cesar França June 28th, 2019 3:17 am
    Please,
    How can I use the text mining functionality of VOSviewer?
    Thank you!
    Cesar Franca
    Reply
  • Trang Pham June 26th, 2019 5:18 am
    Dear all,
    I am very interested in your VOS viewer.
    One issue I am concerned is that does your VOS viewer can run database fromWeb of Science- Thomson (all database) or only Web of Science (Core Collection)?
    Hope to hear from you soon.
    Reply
    • Ludo Waltman June 27th, 2019 10:29 am
      Both are supported, but data from the Core Collection offers additional possibilities. Using Core Collection data, it is possible to create citation-based maps. This is not possible using data from all databases.
      Reply
  • Jessica Bell June 19th, 2019 3:35 pm
    When creating a keyword thesaurus, I'm getting an error that there are the same terms in the label column as in the replace by column and there are not. I didn't receive this error yesterday, but am today. I have no idea what I'm doing incorrently!
    Reply
    • Ludo Waltman June 20th, 2019 11:39 pm
      Please send your file to vosviewer@cwts.leidenuniv.nl and I will take a look at your problem.
      Reply
  • Julea Shaw June 11th, 2019 8:40 pm
    When creating a keyword co-occurrence map, you are able to choose the minimum number of occurrences of a keyword in order to include it in the map. How do you know what the threshold should be to include a term?
    Reply
    • Ludo Waltman June 12th, 2019 7:41 am
      My recommendation is to try out different thresholds to explore the effects on the selection of keywords and on the map that is obtained. In general, it is recommended to create multiple maps based on different parameter choices in order to explore the sensitivity of the results to the parameter choices. In a map, you can see for each keyword how often it occurs. In this way, you know for each keyword whether it will be included or excluded for a specific choice of the minimum number of occurrences threshold.
      Reply
  • Andy Yeung June 6th, 2019 10:39 am
    Fractional counting of authorship: Does fractional counting affect the strength of "link" only, or can it adjust for the actual count when the map.txt is viewed in Excel? I want to check out the adjusted count, which is impossible from Web of Science's Analyze function. Thank you.
    Reply
    • Ludo Waltman June 8th, 2019 10:33 am
      Could you please clarify your question. What do you mean by 'actual count' and 'adjusted count'?
      Reply
      • Andy Yeung June 8th, 2019 3:52 pm
        Actual count is the count obtained by full counting. But I want to see a count that is adjusted by co-authorship. For example, if a paper is authored by both Chinese and USA authors. By full counting, China = 1 and USA = 1. By fractional counting, China = 0.5 and USA = 0.5. Can the values from the latter be found in the map.txt? I know that the link strength changes if I choose fractional counting instead of full counting, but link strength is not what I want.
        Reply
        • Ludo Waltman June 8th, 2019 5:31 pm
          VOSviewer does not provide this type of fractional counting statistics. Fractional counting is available only for the calculation of the link strengths, not for the calculation of the weights of the items.
          Reply
          • Andy Yeung June 9th, 2019 12:39 am
            Thank you for your clarifications!
            Reply
  • Andy Yeung June 6th, 2019 9:55 am
    Hi - Is it possible to search only for publications in WoS that have an PMID? I would like to present visualizations together with data of MeSH descriptors. Hence, I am not interested in publications not indexed in PubMed.
    Ultimately, I want to define a body of literature by MeSH descriptors, get their bibliometric data from WoS (or Scopus), and visualize by VOSviewer. Is that possible?
    Reply
  • Anne Zehoul June 5th, 2019 7:45 pm
    Hello
    I am a first time user of VosViewer, and got stuck in the first step : when I try to select the SCopus csv file it appears a message :
    Error while reading Scopus file 'xx': IOException reading next record: java.io. IOException : (line2) invalid char between encapsulated token and delimiter.
    Do you know how to fix it ?
    tks a lot !
    Reply
    • Ludo Waltman June 5th, 2019 8:23 pm
      This is a bug in the Scopus CSV export. We are in contact about this with the Scopus team at Elsevier. They are working on fixing the bug. As a temporary workaround, you could try to export only selected fields rather than all available fields. In particular, excluding abstracts from the CSV export might help.
      Reply
      • Anne Zehoul June 5th, 2019 11:52 pm
        Thank you Ludo !
        Reply
        • Stephanie September 7th, 2019 5:06 pm
          Hi
          I have the same error message / problem with a Scopus export file. It seems the bug is not fixed yet. Is there another work around by manipulating the csv export file, I'd really like to keep the abstract information?
          thanks!
          Reply
          • Ludo Waltman September 8th, 2019 11:16 am
            Stephanie, could you send your file to vosviewer@cwts.leidenuniv.nl? I will then take a closer look at your problem.
            Reply
            • Marieke Venselaar November 11th, 2019 4:24 pm
              Dear Ludo,
              I have a problem that is related to Anne's and Stephanie's question.
              I have two CSV-files. From both files I exported only limitited fields: Authors,Author(s) ID,Title,Year,Source title,Volume,Issue,Art. No.,Page start,Page end,Page count,Cited by,Link,Affiliations,Authors with affiliations,Author Keywords,Index Keywords,References,Abbreviated Source Title,Document Type,Access Type,Source,EID
              The first file works perfectly, but when I try to select the second file I get the message: "Error while reading Scopus file 'xx': IOException reading next record: java.io. IOException : (line2) invalid char between encapsulated token and delimiter"
              Do you know what I can do to avoid this?
              I tried to make one file out of the two files, but that did not work.
              Thank you in advance!
              Marieke
              Reply
  • PJH June 4th, 2019 9:42 pm
    Is dimensions data downloaded and analyzed in VOSviewer or only in Dimensions.
    Philip
    Reply
    • Ludo Waltman June 4th, 2019 9:47 pm
      You need to download data in the Dimensions web interface. The data can then be provided as input to VOSviewer. We are exploring possibilities for a more convenient connection between Dimensions and VOSviewer.
      Reply
  • Alex June 4th, 2019 6:06 pm
    Hi - Is it possible to automate the uploading of multiple .csv files via the 'create files based on text data' option and the 'read data from bibliographic database files' specifically? And then confirming all the following options? I've consulted section 5.1 and can't see anything, but as I'm not really familiar with command line procedures I may be missing something. So any pointers would be much appreciated.
    Thanks,
    Alex.
    Reply
    • Ludo Waltman June 4th, 2019 9:50 pm
      This is indeed not possible at the moment. Perhaps we can add this feature in the future. However, I am afraid I cannot make any promises. There is a long list of features that we would like to add to the software, but unfortunately our resources are limited. Thanks for making this suggestion!
      Reply
      • Alex June 5th, 2019 9:11 am
        OK, thank you for confirming!
        Best of luck and kind regards,
        Alex.
        Reply
  • PJH June 4th, 2019 2:36 am
    We wrote a bibliometric paper using VOSviewer where the editor asked if we could confirm the results with a second database (we used Scopus). Given that in the field(s) being reviewed (management-education) Scopus coverage is much broader than WOS, I don't really see the point of this. I think it is a request from someone who doesn't really understand the relevant factors.
    In any case, if we can't convince the editor otherwise, am I correct that there is no way to combine databases from Scopus and WOS together into a single analysis? The best we could do would be to run a second search in WOS and then compare the results?
    Thanks.
    Philip
    Reply
    • Ludo Waltman June 4th, 2019 3:25 am
      Yes, you are right, there is no straightforward way in which data from different data sources can be combined.
      By the way, instead of, or in addition to, Web of Science, you could also consider the use of Dimensions.
      Reply
      • Katia Figueroa August 15th, 2019 8:04 pm
        Hello,
        The same problem here... what do you mean by Dimensions?
        Thank you
        Katia
        Reply
        • Ludo Waltman August 15th, 2019 10:56 pm
          See https://www.dimensions.ai/.
          Reply
          • Katia Figueroa August 15th, 2019 11:47 pm
            Ludo
            Thank you for your kind response, I'm looking for an article/book chapter that says: "there is no way to combine databases from Scopus and WOS together into a single analysis" as PJH posted... Could you help me with this.. please.
            Thank you in advance
            Katia
            Reply
            • Ludo Waltman August 15th, 2019 11:52 pm
              I don't think such an article exists. Combining data from multiple data sources is very challenging. Without an extensive bibliometric expertise, it is not possible. However, with a deep understanding of bibliometric data sources and good access to these data sources, there are possibilities to combine data from multiple sources.
              Reply
              • Katia Figueroa August 16th, 2019 12:14 am
                Ludo,
                Do you happen to know about an article where the authors justify using only one data base (WoS or SCOPUS)?
                Best
                K
                Reply
                • Ludo Waltman August 16th, 2019 8:22 am
                  The justification often is not made explicit, and in many cases it is probably pragmatic. Authors have access to one data source and not to another, which explains their choice of a data source. However, there is quite some bibliometric literature studying the strengths and weaknesses of different data sources. This literature could be used to make an informed and justified choice to use a particular data source.
                  Reply
  • Treshani June 3rd, 2019 11:47 am
    Hi Ludo,
    Could you please tell me if there is any possible way to show two terms together when mapping based on text data (from title and abstract).
    For instance, I want 'virtual reality' as a single label on the map, but my map shows 'reality' separately.
    Thanks
    Treshani
    Reply
    • Ludo Waltman June 3rd, 2019 5:13 pm
      If you don't want 'reality' to be included in your map, you can use a thesaurus file to remove it. You can also use a thesaurus file to merge 'reality' and 'virtual reality', if this is what you want.
      When identifying terms in titles and abstracts, VOSviewer identifies the longest possible terms. Hence, if a sentence includes the phrase 'virtual reality', this is counted as an occurrence of the term 'virtual reality', not of the term 'reality'. However, if for whatever reason the term 'virtual reality' is not included in your map, occurrences of 'virtual reality' will be counted as occurrences of 'reality'.
      Reply
  • Pedro Silva May 28th, 2019 7:12 pm
    I have been playing around with VOSviewer and CItNet, trying to distill some insights from a search query which yielded >4900 hits in Web of Science. Among other things, I had VOSviewer generate a keyword map. Obviously, with such a large number of records I got a large interconnected map but only 5 major clusters. That is fine, but I wonder if there any simple way to modify the program so that one can retrieve (from VOSViewer) the identities of the papers which contribute to each cluster (or to a selected number of terms from a cluster. That would be tremendously helpful when preparing,e.g, a systematic review of the literature: using vosviewer to find broad subfiels, then zooming into a specific region of interest to obtain a manageable group of papers.
    Reply
    • Ludo Waltman May 29th, 2019 3:09 am
      There currently is no easy way to accomplish this. However, when you create a map, in the final step of the 'Create Map' wizard, a table is shown listing the keywords to be included in the map. By right-clicking on this table, it is possible to export the document-keyword relations. By combining the document-keyword relations with the keyword-cluster relations (which can be obtained by saving your map in a VOSviewer map file), it is possible to find out which documents contribute to which clusters. In principle, this offers a solution to your problem, but it is not a very easy one.
      Reply
      • Pedro Silva May 29th, 2019 8:04 pm
        I have followed your instructions and see how it can be done. Thanks! The "only" glitch is that papers are identified by number in the document-keyword file, and in a long list of documents it could be difficult to avoid mistakes when selecting them by hand afterwards.
        Do you think it would be possible to include (in a future update) a simpler way to do that?
        Reply
        • Ludo Waltman May 29th, 2019 10:41 pm
          I cannot promise anything, but I hope we will be able to simplify this in the future.
          Reply
  • PJH May 23rd, 2019 4:41 am
    Hi
    I'm looking at the feasibility of a review project and identified around 40,000 relevant documents in Scopus. Scopus would only allow export of 'citation information' for 20,000 documents at a time. So I tried that and got access to a csv file with 20K documents in it with basic info plus citations.
    However, when I ran it to see what kinds of analyses I'd be able to do in VOSviewer, the only analysis available was co-author of authors. Everything else was grayed out.
    What am I missing or doing wrong here? And, can you suggest a way to get this info from Scopus in a reasonably efficient way. Or would WOS be any better?
    Thanks.
    Philip
    Reply
    • PJH May 23rd, 2019 7:08 am
      With some more work I figured one way to deal with this in Scopus (not sure if it's the best).
      1. Identify documents (40,000 [plus)
      2. Save to independent lists (
      Reply
      • Ludo Waltman May 23rd, 2019 9:28 am
        Unfortunately, the functionality in Scopus for exporting up to 20,000 records is of limited use, since a lot of data will not be included in the resulting file. To get all relevant data, you need to use the regular export functionality of Scopus, which supports a maximum of 2,000 records. You may need to find ways to split up your search results in batches of at most 2,000 records (e.g., by making a breakdown by publication year).
        Reply
  • giuseppe vito May 22nd, 2019 7:21 pm
    Dear Ludo,
    I created an Author Keywords map with bibliographic data from Web of Science (limit: 2 occurrences).
    Many Keywords obtained with Vosviewer are not present in those mentioned by the Author in the full text of the paper.
    What can the explanation be?
    The format of the scores in the Vosviewer map files is not numerical, whether I open them with Excel (2019) or OpenOffice (2019).
    Can you help me?
    Thanks in advance
    Reply
    • Ludo Waltman May 22nd, 2019 8:17 pm
      You probably also included KeyWords Plus. These are keywords assigned algorithmically to publications by Web of Science. These keywords need not occur in the publications themselves.
      Reply
  • Atefeh Zolfagharnasab May 20th, 2019 9:02 am
    Hi dear Ludo
    You said in this article "Text mining and visualization using VOSviewer", VOSViewer can use for text mining.
    There was a function In data mining that called "associated rules". Can we call the co-occurrence a branch of associated rules?
    Reply
    • Ludo Waltman May 20th, 2019 9:05 am
      I don't think co-occurrences can be seen as a branch of association rules. Instead, I think co-occurrences provide the basis for deriving association rules.
      Reply
      • Atefeh Zolfagharnasab May 20th, 2019 9:08 am
        Thank you so much
        Reply
  • João Pildervasser May 15th, 2019 3:37 am
    Hi Ludo,
    When creating a text map using data from WoS (journal: "New Phytologist", year: 2016-2019) and selecting the overlay visualization using "avg. pub. year" the year range automatically generated goes from 2017.2 to 2017.5. If I manually set it to 2016-2019 the map becomes monochromatic as if all publications are from mid-2017. What's the reason for that and is there a way to solve it? Thanks
    Reply
    • Ludo Waltman May 17th, 2019 9:18 am
      The color of a term reflects the average publication year of the publications in which the term occurs. For almost all terms, this average publication year will be somewhere in the middle of your period of analysis. This is why in your case the colors are concentrated between 2017.2 and 2017.5. You can click on the color bar to manually change the year range to 2016 and 2019. You will see that the terms then almost all have the same color, making the color coding of little use.
      Reply
  • Dennis Lundberg May 13th, 2019 4:03 pm
    We are moving away from Oracle's Java because of the changed license terms. Instead we are deploying AdoptOpenJDK, which is a distribution of OpenJDK.
    We are unable to start the downloadable executable of VOSviewer on neither Windows 10 nor MacOS X. All we get is an error message saying "Unable to run the Java runtime environment.
    What can we do to run VOSviewer on OpenJDK? Is there a configuration option that we can use to tell VOSviewer which version of Java to use?
    Reply
    • Ludo Waltman May 13th, 2019 9:22 pm
      Perhaps you could try to run the VOSviewer JAR file, which can be downloaded from http://www.vosviewer.com/download as 'VOSviewer 1.6.11 for other systems'. Please let me know if this doesn't solve your problem.
      Reply
      • Dennis Lundberg May 14th, 2019 4:21 pm
        Thank you!
        That worked for me.
        Reply
Build on Applepie CMS by Waltman Development