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
  • John Paul SanGio September 20th, 2018 4:55 pm
    Thank you for offering this excellent resource.
    I am constructing a co-authorship map.
    I would like to display all authors with whom I have ever published.
    My current map (generated by default settings) contains only those authors with whom I have published a number of times.
    I would appreciate guidance on how to display all co-authors.
    John Paul
    Reply
    • John Paul SanGio September 20th, 2018 8:34 pm
      VOSviewer Staff:
      I solved the problem (found the options with the Web of Science input format).
      There is no need to respond.
      John Paul
      Reply
  • Ana September 11th, 2018 12:42 am
    Good night,
    Could you please explain me the meaning of "total link strength"?
    I am using VosViewer to do a co-citation analysis (cited references) and a column named "total link strength" appears (next to citation) ???
    Thank you very much!
    Best regards.
    Reply
    • Ludo Waltman September 11th, 2018 7:30 am
      This is explained in Chapter 2 of the VOSviewer manual. If this explanation is insufficient, then please let me know.
      Reply
      • Ana September 11th, 2018 10:40 am
        Hello Ludo,
        Thank you for your answer!
        I already had the Vosviewer manual, but I was asking because in the manual it refers to:
        bibliographic coupling links
        co-authorship links, and
        co-occurrence links

        But it is not clear to me in the case of co-citations (refrerences), whats the meaning of "total link strenght"

        Hoping you me can help me.

        Thanks a lot!
        Regards.
        Reply
        • Ana September 11th, 2018 6:49 pm
          Thanks for everything. I found the information!
          Greetings.
          Reply
  • Breno Prata September 9th, 2018 4:31 am
    Good night,
    I am trying to create a web integration between VOSViewer and Gephi, I want it all to work as simple as possible, I want to import a csv file from scopus into VOSViewer and generate a map, after that export that map to a supported Gephi format. But I can't quite find that code on the JAR source file, is that any way to help me find those specific code, it would really make a diference if I managed to unite these two tools :)
    Reply
    • Ludo Waltman September 9th, 2018 9:03 am
      The source code of VOSviewer currently is not available. The software is not open source. We plan to release the source code within the coming months.
      Reply
  • Hai Giang Ha September 7th, 2018 9:44 am
    Dear Mr. Ludo Waltman,
    Could you please tell me how I can explain the relationship of items in a cluster? How was the color of a cluster assigned? And Is there anyway if I want to change the position of one item?
    I need that information to explain in my report.
    Thank you so much and I am looking forward to hearing from you.
    Best regards,
    Hai Giang Ha
    Reply
    • Ludo Waltman September 7th, 2018 7:25 pm
      Your question is a bit general. Some answers are provided in the VOSviewer manual and in the book chapter available above. If these answers are insufficient, then please let me know.
      Reply
  • Philip Hallinger August 28th, 2018 5:32 pm
    Great program.
    I have not had any issues up to now. However when I upload a scopus file, I keep getting an error message that there is an incorrect number of rows in record 367. I went and deleted record 367 in scopus. Now it says the error is in record 368.
    What's the problem and the solution, please.
    Reply
    • Ludo Waltman August 29th, 2018 7:43 am
      Could you please send me your file (vosviewer@cwts.leidenuniv.nl)? I will then take a look at it and offer advice on this issue.
      Reply
      • Philip August 29th, 2018 8:19 am
        Ludo
        Thanks for the prompt response. Although I am an absolute newbie at this I found a workaround for the problem.
        The database was only 700+ documents, so I exported 200 at a time to 4 files and uploaded the four files. it worked. So although I don't know what the 'problem' actually was, I don't have it anymore and I'm a happy guy.
        So thanks again.
        Reply
  • Alberto August 23rd, 2018 10:22 am
    Hi! Thank you for your work on VOSviewer.
    I'm wondering if it is possible to create a co-citation keyword map based on a personal library of papers.... I mean, not taken from scopus or wos, but, for example, from a .bib file. Keywords are there, so I'm wondering if a way to do that actually exists.
    Thanks!!
    Alberto
    Reply
    • Alberto August 23rd, 2018 10:39 am
      I did it with .RIS file very easily from Endnote! Thanks!
      Reply
  • Robyn August 22nd, 2018 6:43 pm
    Hello, I'm working on cleaning up a small set of citations using a thesaurus file. I've reviewed the sample thesaurus files provided, and taken the suggestions in the comments made previously of creating a tab delimited file using Excel. However, when I create my visualization, instead of replacing the terms outlined in the thesaurus file, it simply removes them from the visualization altogether.
    It was my understanding that the "replace by" column would allow me to collect together the alternate spellings/ plurals into on term - not remove the terms entirely. Have I misinterpreted the purpose? Is there a strategy you would suggest to resolve this issue?
    Many thanks!
    Reply
    • Ludo Waltman August 22nd, 2018 9:59 pm
      Your interpretation is fully correct. Could you please send your thesaurus file to vosviewer@cwts.leidenuniv.nl? I will then take a look at the file and try to offer some advice.
      Reply
    • Alberto September 9th, 2018 9:50 pm
      Hi Robyn,
      could you please tell me where I can find the thesaurus samples?
      Thanks!
      Reply
  • Giuseppe Vito August 19th, 2018 5:37 pm
    Dear Ludo,
    Thanks for the reply.
    The txt files I intend to make available on-line are those generated by Vosviewer in the analysis of bibliographic coupling, cocitation, etc. (... map.txt and ... net.txt) of publications of the department I belong (exactly as discussed in section 5.2 of the manual). Currently the aim is to divulge bibliometric maps to colleagues of the department, subsequently I intend to integrate some of my publications with links to the related Vosviewer maps.
    In the meantime I solved the problems with github, which constantly flagged my account by mistake: I'm sorry for the request made to you.
    On this occasion:
    using excel to read a file ... map.txt (eg. bibliocouplingmap.txt), the format of the individual cells of the columns weight and score is not unique and I cannot transform them into one numerical format as I would like to perform further reflections.
    is it possible to generate a network file in full format (pages 38-39 of the manual version 1.6.8) instead of sparse format?
    Thank you very much for the attention you will dedicate to me.
    Reply
    • Ludo Waltman August 19th, 2018 5:47 pm
      Your Excel problem probably relates to the regional settings of your operating system. The easiest way to solve the problem then is to change the regional settings. You need to make sure that a period, not a comma, is used as the decimal separator.
      Saving a network file in the full format is not supported by VOSviewer. VOSviewer can open network files in two formats, both the sparse and the full format, but it always saves these files in the sparse format.
      Reply
      • Giuseppe Vito August 19th, 2018 5:55 pm
        Thanks for the immediate reply, I will try to change the settings in Excel and I will let you know
        Reply
  • Giuseppe Vito August 16th, 2018 5:43 pm
    Dear Ludo and Nees Jan, first of all my congratulations for your work and my best wishes for future developments. I encounter problems in making txt files available online. I have used github but I am not fully satisfied with this software. Could you kindly suggest alternative ways of doing this? Best Regards. Giuseppe Vito, Parthenope University, Naples, Italy
    Reply
    • Ludo Waltman August 19th, 2018 1:23 pm
      I think I need a bit more information to answer your question: What kind of text files do you want to make available online, and for what purpose? Do you mean that you are trying to make a map available online, as discussed in Section 5.2 of the VOSviewer manual? And why exactly are you 'not fully satisfied' with GitHub?
      Reply
      • Giuseppe Vito August 19th, 2018 5:33 pm
        Dear Ludo,
        Thanks for the reply.
        The txt files I intend to make available on-line are those generated by Vosviewer in the analysis of bibliographic coupling, cocitation, etc. (... map.txt and ... net.txt) of publications of the department I belong (exactly as discussed in section 5.2 of the manual). Currently the aim is to divulge bibliometric maps to colleagues of the department, subsequently I intend to integrate some of my publications with links to the related Vosviewer maps.
        In the meantime I solved the problems with github, which constantly flagged my account by mistake: I'm sorry for the request made to you.
        On this occasion:
        1) using excel to read a file ... map.txt (eg. bibliocouplingmap.txt), the format of the individual cells of the columns "weight " and "score " is not unique and I cannot transform them into one numerical format as I would like to perform further reflections.
        2) is it possible to generate a network file in full format (pages 38-39 of the manual version 1.6.8) instead of sparse format?
        Thank you very much for the attention you will dedicate to me.
        Reply
  • Fernanda August 13th, 2018 5:18 am
    I am trying to download the VOSviewer for windows at this site. Althoug, my computer detected a virus
    The name is Trojan:Win32/Vigorf.A
    Is it safe to make the download here?
    Is there any problem with the website?
    Tkx
    Reply
    • Ludo Waltman August 13th, 2018 1:24 pm
      A number of VOSviewer users have reported this problem. We have investigated the problem in detail. The software is safe, and the threat warning given by Windows Defender is incorrect. We have added a brief explanation on the Home page and also on the Download page.
      Reply
      • Ludo Waltman August 19th, 2018 1:24 pm
        We have been in contact with Microsoft's Windows Defender team and the problem has now been solved.
        Reply
  • Ranbir August 10th, 2018 11:58 am
    Dear Friends
    If anybody can help me to answer
    Out of ---first author co-citation data and all-author co-citation data, which one is used by VOSviewer ?
    Your advice can very helpful for me
    Reply
    • Ludo Waltman August 10th, 2018 12:11 pm
      In the case of Web of Science data, only the first author is considered (since information on other authors is not available in Web of Science data). In the case of Scopus data, all authors are considered.
      Reply
      • Ranbir August 10th, 2018 12:39 pm
        Thanks a lot Dear Ludo Waltman
        Thanks again
        Reply
  • Anthony Molisani July 26th, 2018 5:45 pm
    Hi Dr Waltman,
    I really appreciate the text mining features of VOSviewer, In doing the analysis, I would like to further explore the publication metrics associated with the clusters of words. To do this, I need to know which publications are associated with each term. I'm having trouble finding such a way to identify/export theses relationships between publications-mapped terms. Any advice would be incredibly helpful!
    Reply
    • Ludo Waltman July 26th, 2018 6:16 pm
      In the last step of the 'Create Map' wizard, a table is presented that shows the terms that will be included in your term map. It is possible to right-click on this table. A context menu will then appear that offers the possibility to save the links between terms and documents. This seems to be exactly what you are looking for. Let me know if you need further information.
      Reply
      • Anthony Molisani July 26th, 2018 7:50 pm
        Hi Dr Waltman,
        That is perfect. Thank you!
        Regarding document id numbers, I imagine they are assigned based on the row within the upload document. How are they assigned when using multiple files when uploading records into VOSviewer?
        Thanks, again. This is extremely helpful!
        Reply
        • Ludo Waltman July 26th, 2018 8:00 pm
          When you have multiple input files, the document IDs are dependent on the order in which your files are processed by VOSviewer, and this order is not visible to the user. In this case, it is therefore best to manually merge your files into a single input file. The document IDs then correspond in an unambiguous way to rows within your input file. Note that, if your input file includes a header line (as is the case for Web of Science and Scopus files), the first document is located on the second line in your file and the ID of this document is 1. The document on the third line has ID 2, and so on.
          Reply
          • Anthony Molisani July 26th, 2018 9:06 pm
            Excellent. Thank you very much!
            Reply
  • Hai Giang Ha July 15th, 2018 12:17 pm
    Dear Mr. Waltman,
    I am now using this wonderful tool for bibliometric analysis. The records download from the Web of Science are quite large, about 250,000 records. When I insert the records, the system informed that I need to extend the memory. I have read the manual, yet, I could not find VOSviewer JAR file as well as the command line.
    Could you please tell me where I can download the VOSviewer JAR file and the command line?
    I am looking forward to hearing from you.
    Thank you so much.
    Hai-Giang Ha
    Reply
    • Ludo Waltman July 15th, 2018 1:39 pm
      The JAR file can be downloaded from the VOSviewer website. It is referred to as the VOSviewer version for other systems (i.e., not for Windows or for Mac systems). In principle, the Windows and Mac versions of VOSviewer should already make use of most of the internal memory available on your system, so I am not sure whether the use of the JAR file will solve your problem. I am wondering what kind of map you would like to create based on your 250,000 records. If you could provide me with some more details, I may be able to offer some further advice. It would also be helpful for me to know you much internal memory you have available on your system. (By the way, the amount of internal memory that is made available to VOSviewer can be found in the bottom part of the 'About' dialog box in VOSviewer.)
      Reply
      • Hai Giang Ha July 15th, 2018 5:24 pm
        Thank you so much for your quick reply.
        The amount of internal memory that I found in the 'About" dialog box is 'Memory usage: 37/ 746 MB'.
        I want to create an (author) keywords co-occurrence and countries co-authorship network.
        Reply
        • Ludo Waltman July 15th, 2018 7:17 pm
          746 MB is a small amount of memory. This may cause problems even for data sets much smaller than yours. How much internal memory does your computer have in total? Do you have many other programs that are running simultaneously with VOSviewer?
          Reply
          • Hai Giang Ha July 16th, 2018 5:17 am
            I have installed VOSviewer on my other computer, driver D. Its volume is 537 GB, and it has 515 GB free. The system information: RAM 8GB. Memory usage this time somehow is better '44/ 989 MB'. However, I still can not create the maps with that 250,000 records. When I used VOSviewer, I closed almost programs but MS Word, MS Excel, and Chrome.
            Thank you
            Reply
            • Ludo Waltman July 16th, 2018 9:34 am
              Thanks for the additional information. Based on this information, it seems that you are probably working with a 32-bit version of Java. In order to get access to more internal memory, you need to use a 64-bit version of Java. This means that you need to update Java on your system,. My recommendation is to update to the most recent 64-bit version of Java 8. (Do not use Java 9 or 10.)
              Reply
              • Hai Giang Ha July 19th, 2018 10:57 am
                Thank you so much for your suggestion.
                After reinstalling Java 8 64-bit for Window, it really worked. When I checked "About", it stated "Memory usage: 109/4132MB". I uploaded the 250,000 records, it loaded 99% and then stopped (before it was only 32%).
                Do you have any solution for me?
                Reply
                • Ludo Waltman July 19th, 2018 11:49 am
                  I am glad to hear that 64-bit Java works better. It seems that you still don't have enough memory to process such a large data set. I think you need to either reduce the size of your data set or find a computer with more internal memory. I hope this is possible for you.
                  Reply
                  • Hai Giang Ha July 22nd, 2018 8:08 pm
                    Thank you for spending your precious time to help me.
                    I install VOSviewer on other computer and it worked.
                    I do appreciate what you and your team have done for the global scientific research.
                    Reply
  • Fatima Noora July 6th, 2018 8:55 pm
    Dear Mr Waltman,
    Could you kindly advise on the way to create Thesaurus file? (the simplest possible, please)
    I am stuck because of the many synonyms, spelling differences and abbreviated terms I've in my Scopus data base.
    Thanks in advance,
    Fatima
    Reply
    • Ludo Waltman July 6th, 2018 11:07 pm
      When you download VOSviewer, you will obtain a ZIP file that includes two examples of thesaurus files. Please take a look at these examples. Let me know if you have any questions based on these examples.
      Reply
      • Fatima Noora July 6th, 2018 11:12 pm
        I already did! they are terms and authors.
        My question is how to create a file like this?
        Reply
        • Ludo Waltman July 6th, 2018 11:22 pm
          It is easiest to create such a file using a spreadsheet program, for instance MS Excel. Make sure to save your thesaurus as a tab-delimited text file.
          Reply
          • Fatima Noora July 6th, 2018 11:48 pm
            Well noted!
            Thank you so much.
            Reply
  • William Sanger July 5th, 2018 7:23 pm
    Hi M. Waltman,
    First of all, thanks for such amazing tool VOSviewer is. I've been using it for the past 2 years to visualize batchs of academic papers, which was great.
    I have a question. I cannot do usual visualizations with VOSviewer using Web of Science exportation tool (5K). Now, when I export papers, it only shows the title, the author and the source. I do not have access to the keywords anymore. The same goes with "save to other file formats".
    Do you have any insight about such situation? If so, do you know any alternative?
    Best regards, and thanks a lot again,
    William
    Reply
    • Ludo Waltman July 5th, 2018 7:35 pm
      Using the 5K option, files exported from Web of Science do not provide full publication records. To obtain full publication records, you need to use the ordinary export feature of Web of Science, where downloads need to be done in batches of at most 500 records. I am afraid this is the only solution. When you use the ordinary export feature of Web of Science, it is possible to download full publication records. VOsviewer will then be able to create keyword co-occurrence maps and other types of maps that cannot be created based on incomplete publication records.
      Reply
      • William Sanger July 9th, 2018 2:53 pm
        Thanks a lot,
        Have a great day,
        W.
        Reply
  • Shivam June 28th, 2018 8:59 pm
    Dear Sir,
    can a VOS viewer be done for the Data downloaded from google scholar using publish and perish tool in csv format ?
    or what other methods can be used to do an analysis for google scholar data ?
    Reply
    • Ludo Waltman June 28th, 2018 10:06 pm
      Yes, it is possible to work with Google Scholar data in VOSviewer. This can be done by downloading data from Google Scholar using Publish or Perish and by saving the data in a RIS file. Unfortunately, the possibilities for analyzing Google Scholar data in VOSviewer are limited. In particular, it is not possible to create citation-based maps based on Google Scholar data.
      Reply
  • Ilenia June 26th, 2018 7:25 pm
    Hello,
    I would like to use VosViewer for my research, however I am having issues in uploading files from Scopus.
    I did my research on scopus and I exported in a CSV file all information that Scopus allows me to export i.e. citation, bibliographic information, etc. When I open the downloaded file I can see all the details of the papers I selected in my search. When I upload the file in VosViewer and I try to create a term map VosViewer displays this message: "Error while reading Scopus file: file is empty". I do not understand what is the problem since I have not changed anything in the downloaded file. Could you please help me with this issue?
    I thank you in advance.
    Best regards,
    Ilenia
    Reply
    • Ludo Waltman June 26th, 2018 10:20 pm
      Could you please send your file to vosviewer@cwts.leidenuniv.nl? I will then take a look at your problem.
      Reply
      • Ilenia June 27th, 2018 1:10 am
        Hi Ludo,
        I thank you for your reply but it seems that I sorted out the problem. I had to make sure that the content in the first cell of the first row was split in different colomns. Once that I did this, the software is now working.
        Best regards,
        Ilenia
        Reply
  • Jasmijn June 25th, 2018 1:55 pm
    Hi there,
    Could you give me some advice on the following?
    I learned about VOSviewer through a colleague and wonder whether the software is suitable for a literature review I'm currently conducting:
    After a systematic search I've ended up with a small dataset of 45 suitable references. I notice an immense range of operationalizations for the same concept that are used by these references and want to delve further into this issue. I want to visualize the different measurements, group them and show how many references use the same measurement. I think there are several islands of measurements that are used by the same authors and I would like to map this. I then would like to see whether the references cite each other and how many references build bridges to other islands of measurements.
    Would it be possible to create the network maps I'm looking for in VOSviewer or does the software only produce maps based on references as data? I guess that my research interests would require me to create and import data (measurements) that I would have to construct myself. Based on what I read on VOSviewer so far, I am not certain whether the software is suitable for this specific literature review.
    Thank you very much for considering my request.
    Best wishes, Jasmijn
    Reply
    • Ludo Waltman June 25th, 2018 6:38 pm
      Although I don't fully understand what kind of visualization you would like to create, I do think it is possible to create such a visualization using VOSviewer. However, this will require an in-depth knowledge of VOSviewer, since you will need to do considerable preprocessing of your data to provide the data to VOSviewer in the right way. If you do not yet have a substantial experience in working with VOSviewer, then I think creating the visualization you have in mind will be quite challenging.
      Reply
      • Jasmijn June 26th, 2018 10:23 am
        Thank you very much for your swift reply! This is helpful in deciding what to do. Best wishes, Jasmijn
        Reply
  • Lara June 23rd, 2018 12:32 am
    Hi there,
    I've been using VOSViewer for a couple of years now, and I really like what you can do with it. However, I find myself unable to use it for graphic analyses of co-citations (i.e., nb of times specific papers get cited in a series of references). I've tried to upload databases with .RIS and .CSV extensions but I can't seem to get it right: I never get the option to analyse co-citations (only co-authorship and co-occurences). What is the right format to run a bibliomeric analysis of co-citations? If the source should be .CSV, what should my table look like?
    Thanks for your advice,
    Lara
    Reply
    • Ludo Waltman June 23rd, 2018 12:42 am
      Citation-based analyses indeed are not supported for RIS files. This is because RIS files do not contain reference data in a standardized format. Citation-based analyses are supported only for Web of Science files, Scopus CSV files, and Crossref JSON files.
      Reply
      • Lara June 23rd, 2018 12:51 am
        Thanks a lot for your prompt response! Ok, it makes sense. However, none of these research databases (WoS, Scopus and Crossref) provides the exact list of refs. Is it possible to build the file manually? Is there a sample I can follow?
        Reply
        • Ludo Waltman June 23rd, 2018 7:43 am
          I am not entirely sure what you mean by "none of these research databases (WoS, Scopus and Crossref) provides the exact list of refs". What do you mean by "the exact list of refs"? Anyway, in order to have an example, just download a file from Web of Science or Scopus and use that file as an example. Please keep in mind that you need to follow exactly the file format used by Web of Science or Scopus, including the format used for specifying references. Otherwise VOSviewer will not be able to process your data.
          Reply
          • Lara June 23rd, 2018 4:13 pm
            Ok, noted. Thank you for your help!
            Reply
  • Allison June 22nd, 2018 6:57 pm

    How I can download a CSV file of a map I have created in VOS viewer? I would like to download the data used to create the map and use this to calculate some statistics.
    Thank you!
    Reply
    • Ludo Waltman June 22nd, 2018 8:15 pm
      Use the 'Save' button on the 'File' tab to save your map in a map and network file. These are plain text files with tab-delimited columns. Please see the VOSviewer manual for more details on the map and network file formats.
      Reply
  • rinrin June 17th, 2018 2:26 am
    Greetings,
    The VOSviewer manual refers to txt files (says "see....txt file) throughout. Where are these txt file example located so that I can view them?
    Reply
    • Ludo Waltman June 17th, 2018 9:02 am
      When you download VOSviewer from the website, you will obtain a ZIP file that includes these text files.
      Reply
  • Andy Yeung May 31st, 2018 1:12 am
    Hello everyone,
    I conducted a bibliometric study using VOSviewer. Journal reviewer's main concern is the use of a single database, Scopus. S/he mentions that I should search multiple database to avoid missing some highly-cited papers not indexed in Scopus. Suppose I also search WoS, then a lot of papers will have two distinct citation counts... how should I merge them altogether for the overall analysis, e.g. top-100 list, country and journal performances?
    Many thanks and best wishes,
    Andy
    Reply
    • Ludo Waltman May 31st, 2018 8:03 am
      Merging two different data sources is highly challenging. You need to have an extensive expertise on the data formats used by the different data sources in order to be able to merge them. My recommendation is not to take this approach. It probably will be too difficult and too complex. Although I don't know the details of your project, I think it is best to use only the Scopus database. In almost all cases, publications that are in Web of Science are also in Scopus, especially highly cited publications, so I do not expect that adding Web of Science data has a substantial effect on your study.
      Reply
      • Stephan De Spiegeleire June 3rd, 2018 2:39 am
        Hmm. Disappointing answer. I think the issue raised here is a quite important one - we're running into similar issues in a large project. One of the ways in which we hope to solve this problem is via Zotero and the update to the DOI plugin that Brenton Wiernik is planning to make that will allow the user to automatically add DOIs to Zotero items through the CrossRef OpenURL API. If VOSviewer could include something similar, that would really be great. 
        Another option - used by VantagePoint and CiteSpace, for instance - might be to use fuzzy logic, with a degree of conservatism that the user can define, to generate an initial list of potentially matching authors (or organizations, etc. -whatever fields have to be 'cleaned'). The user then review the list and either approve or have the ability to move one or more of the entries that are 'preliminarily matched' to where they should be. 
        Finally, we are also looking into more generic tools like OpenRefine (formerly Google Refine), which might also do the trick.
        But so I certainly hope VOSviewer will look a bit deeper into this, as it is otherwise a really powerful and relatively intuitive tool! 
        Reply
        • Ludo Waltman June 3rd, 2018 10:03 am
          I think it would not be fair to give the impression to VOSviewer users that the problem of merging data from Web of Science and Scopus can be solved in an easy way. We need to acknowledge that there is no easy solution. The problem of course can be solved, but this requires considerable technical skills. Most VOSviewer users do not have these skills.
          Also, merging data from Web of Science and Scopus is of limited interest, since to a large extent Web of Science is a subset of Scopus, at least under the assumption that Web of Science is to be understood as Web of Science Core Collection and that the analysis focuses on publications going back at most 20 years.
          Because of the above two reasons, from a practical point of view, my recommendation to the typical VOSviewer user is to work with a single data source.
          Reply
  • Ghaliamus May 29th, 2018 1:37 am
    Hello,
    Many thanks for your appreciated work behind such amazing bibliometrics tools.
    Let me please ask about two possible explanations by which I can understand my resulted network analysis. To make my question clear, if you don't mind, I will use the produced network figure, you put on your VOSviewer tutorial cover, as example. Let us assume the network is based on WOS data. Let us assume the network is a network of sources (journals, conferences, etc ..) and the links between such sources (nodes) are based on the citations go and back between every two nodes. The two needed interpretations, if you can provide please, are:
    (1) The clusters: what is the explanation of having two sources such as Nature and Science in one cluster? or having PLoS ONE in a different cluster? I read your tutorial and read about the clustering techniques. However, my question could be considered a question from layman who wants only to understand what WOS data were important in putting two sources in one cluster or in different clusters? Was the data the keywords? the title? only the number of citations?
    (2) The distances between any two nodes, regardless they are in same cluster (colour) or in different cluster: what is the explanation of having two nodes near each other or far from each other? what makes Journal of Geoghysical Research near Sensors journal but not near The Plant Journal? Again, was the reason the citations number or the documents content (such as abstracts) or the authors?
    I'm sorry if I'm asking about the obvious. With your appreciated kindness, I wish you can give me such explanation. Many thanks and my best wishes.
    Reply
    • Ludo Waltman May 29th, 2018 6:33 am
      Both the clustering (i.e., the coloring of the items) and the layout (i.e., the positioning of the items in the two-dimensional space) are based exclusively on the underlying network data, so the citation links between journals in your example. Hence, journals that are in the same cluster tend to be strongly connected to each other by citation links. Likewise, journals that appear close to each other in the layout tend to be strongly connected to each other by citation links.
      Reply
      • Ghaliamus June 5th, 2018 3:36 pm
        Hello Mr Ludo, many thanks for your appreciated response and its explanation. I beg your patience with me trying to learn. Before I write this reply I went back to your references focusing more on understanding the VOS algorithm and comparing it to other techniques. Generally, what I got was that the optimization with VOS is to maximize the distance in the map between distinct items and minimize the distance in the map between similar items. Until this point, I can understand what is the meaning of having two journals near each other in a produced map. However, I couldn't figure from my reading the clustering interpretation. And what I got from your kind response that if two items are in one cluster they are relevant and if two items near each other in the map regardless of their clusters they are also relevant. Good. What I want to distinguish is in the following case:
        A map with several items (T) and clusters (C). Focusing on T1, T2 and T3 where T1 and T2 belong to Ca and T3 belong to Cb. According to the distance, however, T2 is near T3 while T2 is far from T1.
        My question is:
        Which one of the following statement is true:
        Statement 1: T2 is relevant to T1 more than T3.
        Statement 2: T2 is relevant to T3 more than T1.
        Statement 3: cannot quantify the difference relevancy of T2 with T3 and T2. Enough to say it is relevant to T2 and T3 without any differentiation.
        Reply
        • Ludo Waltman June 7th, 2018 9:46 am
          Good question! I would choose statement 3. In your example, the layout and the clustering provide contradictory information. In itself, this is understandable. The layout and the clustering both provide a high-level summary of a complex data set. In summarizing the data, some information is getting lost, and this is why the layout and the clustering do not always provide consistent results. When this happens, the implication is that it is not possible, based only on the layout and the clustering, to draw a clear conclusion on the relatedness of the different items.
          Reply
          • Ghaliamus June 10th, 2018 10:27 pm
            Hello Mr Ludo, many thanks for your patience and explanation, your time is appreciated. If I got your point, it seems that the situation I presented above requires the input of a domain expert while the layout and clustering were not enough. Thank you again for the answer. Good luck.
            Reply
  • Sushma Dhital May 2nd, 2018 4:28 am
    Hello, I am having a trouble. I used web of science data to analyze the citation by documents in Vosviewer. But the xcel has citations written as 'total citations' number greater than what I see in Vosviewer. Why is there the difference? What citation by documents refers to in vosviewer. for instance, i found the result 4081 total citation in xcel and 3891 in vosviewer.
    Reply
    • Ludo Waltman May 2nd, 2018 5:29 pm
      Could you please explain your question in more detail? A detailed step-by-step explanation of the analysis that you are carrying out would be helpful. At the moment, I am afraid I don't fully understand your question.
      Reply
  • Shivam Mani Tripathi April 29th, 2018 5:19 pm
    Dear Sir/ Madam,
    As part of a research article,i am doing a syntometric analysis of google svholar data for the author BS Kademani.
    i have downloaded the details of all his papers including his titles,Coauthors,citations,Source Journals and year of publication.
    i Wanted to do a VOS viewer for this set of data,but as this format is not being supported by the software,how do i proceed further ?
    i wanted to find out the coauthorship network and words network fro this data set.
    Kindly help.
    Thanks
    Shivam Mani Tripathi
    Student Msc
    South Asian University
    New Delhi.
    Reply
    • Ludo Waltman April 29th, 2018 7:01 pm
      My recommendation is to use the Publish or Perish tool to download data from Google Scholar and to save this data in a RIS file. The RIS file can then be processed by VOSviewer.
      Reply
  • d April 28th, 2018 11:35 pm
    Is it possible to use Vosviewer to analyze data collected from JSTOR?
    Reply
    • Ludo Waltman April 29th, 2018 12:04 am
      VOSviewer doesn't provide support for JSTOR data. You need to take care yourself of converting the data to a file format that is supported by VOSviewer.
      Reply
  • Angela April 25th, 2018 12:05 pm
    Hello.
    I am new at VOSviewer and I am doing a bibliometrics analysis about a specific theme on psychology. I have done search in three data bases (WOS, ERIC and MedLine) but I do not know if it is possible import the data of these three data bases and how could I do it. For example: Can I create a excel file with all columns of WOS File and add rows with the information of the others articles that I found in others data bases?
    Reply
    • Ludo Waltman April 26th, 2018 10:17 am
      As you probably know, VOSviewer offers direct support for Web of Science data and for PubMed data (using the Medline file format). ERIC data perhaps can be saved in the RIS file format. RIS files are supported by VOSviewer. Combining different data sources is always challenging. It requires an in-depth understand both of the various file formats and of VOSviewer. If you are interested in producing term co-occurrence maps, different data sources can be combined by creating a VOSviewer corpus file. If you are interested in producing other types of maps, then unfortunately there is no easy way to combine different data sources. Converting data to the Web of Science file format in principle is possible, but in practice this is rather challenging. I do not recommend this approach, unless you are very experienced in advanced data processing.
      Reply
  • Maria P. April 17th, 2018 12:01 pm
    I am interested in making a co-authorship map with data downloaded from SCOPUS. I have seen this option is not supported for the current vosviewer version. Is there a way of downloading the 1.6.5 version? The reason I want to use SCOPUS data is that I can easily download all articles from the institution where I work from SCOPUS (we have an scopus institution ID, and all our articles are associated) but not with Web of Sciences (researchers write the affiliation address in too many different ways). thanks!
    Reply
    • Nees Jan van Eck April 19th, 2018 10:32 pm
      VOSviewer version 1.6.5 can be downloaded using the following links:
      Windows: http://www.vosviewer.com/download/f-y2t2.zip
      Mac: http://www.vosviewer.com/download/f-y2v2.zip
      Other systems: http://www.vosviewer.com/download/f-y2u2.zip
      Reply
  • Amin April 6th, 2018 1:27 pm
    Dear Ludo,
    I am having difficulty with the export from web of knowledge to VOSViewer. When I choose "export to other formats" --> "records (1-269)"--> "Full record and cited references" --> "tab-delimited (win)" --> "Send"
    Then in the software I try to "create a map based on the bibliographic data" --> --> in the web of science tab I choose the text file---> citation -document
    Then instead of the number of 269 records I end up with 219.
    I tried to extract info to Endnote and indeed the number of records are 219!
    I am puzzled of why from 269 records in Web of knowledge only 219 show up in the export process.
    If I export " "Full record" then it comes with 269 records.
    However, then I cannot make a citation document map. Do you have any suggestion?
    Reply
    • Mihaela April 6th, 2018 10:52 pm
      Hello Amin,
      I recently tried to work with web of science and I am having similar problems with "Full record and cited references". It either exports a empty tab-delimited, one were first column headers are "null", completely missing headers, or in the happy event were everything is ok some records missing. I think its a wos problem at export. I managed to work around by exporting batches of 6-14 articles at a time, but its hard work and i only have around 200 articles. I cant imagine doing the same with a large sample of articles.
      To note that a few months ago everything worked ok.
      Reply
      • Amin April 7th, 2018 7:16 am
        It also worked for me few months ago. I tried to extract them in smaller batches of 50. I would do it with what you suggested. Many thanks.
        Reply
        • Ludo Waltman April 7th, 2018 7:22 am
          I'll do some further investigation of this problem. As a temporary workaround, my suggestion is to export data from Web of Science in plain text format instead of tab-delimited format. VOSviewer also supports the plain text format.
          Reply
          • Amin April 7th, 2018 9:43 am
            Dear Ludo
            I I tied and an error returned
            Error while reading web of science file File is invalid
            Reply
            • Ludo Waltman April 7th, 2018 10:28 am
              Amin, Mihaela, could you please send some examples of invalid files to vosviewer@cwts.leidenuniv.nl? This will help me to investigate the problem. Thanks!
              Reply
              • Amin April 9th, 2018 11:33 am
                Dear Mihaela and Ludo,
                I tried to export records in smaller batches of 10. However, it seems the problem still persists. I figured out that some of the records have a value of (null). So, I miss some of the information on the citation data for some of the results. The only thing that I can do is to skip that particular record. Michaela: Can you confirm that this also the case for you?
                Reply
                • Ludo Waltman April 9th, 2018 12:02 pm
                  Brief update: It is clear now that the problems occur both in tab-delimited files and in plain text files. Indeed, as indicated by Amin, one of the problems is the appearance of 'null' strings in the files. Another problem relates to records that are missing in the files. I am in contact with the Web of Science team and they are investigating these problems.
                  Reply
                  • Amin April 9th, 2018 12:33 pm
                    Dear Ludo,
                    In my search I found out that only two records are missing (as null). However if I want to export at once then from 269 only 219 records would be there. Somehow it seems that Web of Knowledge export function does the job in batches. So, if in one of the batches there is a NULL record then the whole batch comes at NULL. In my search records 74 and 210 are NULL. I cab send you the search phrase if needed.
                    Reply
                    • Mihaela April 11th, 2018 2:11 am
                      Sorry for not giving feedback, i was out of town for hollyday. I will send some examples to you Ludo. Amin i suspect the same, batches download. I started with larger batches and then downscaled. In some parts i had to take the articles one by one, and in the end i am only missing around 10 articles out of 250. But i had the whole list of titles downloades before so i searched after the title.
                      Reply
                      • Ludo Waltman April 19th, 2018 7:58 pm
                        Clarivate Analytics has informed me that the problem with the Web of Science export has been fixed. If you still experience difficulties, please let me know.
                        Reply
                        • Mihaela April 19th, 2018 8:01 pm
                          Thank you Ludo,
                          I will try as soon as possible.
                          Reply
  • Patrick Zawadzki March 31st, 2018 5:50 pm
    Hello Ludo! Congrats for this great software, you are doing a great job.
    We just started to use it and found problems on how to do some interpretation.
    When using co-ocurrence of authors keywords some networks shows keywords from different clusters, if the cluster is a correlation between keywords why this happens?
    Cheers
    Reply
    • Ludo Waltman March 31st, 2018 10:03 pm
      I am afraid I don't understand your question. What do you mean by "some networks shows keywords from different clusters"?
      Reply
      • Patrick Zawadzki April 17th, 2018 1:43 am
        For example, we found a network with three nodes, first and second nodes are inside red cluster, and, the third node is inside blue cluster. Our analysis is looking for associated keywords in a specific new area, it is better to use the cluster ou the network results as parameter to decide better association?
        PD: sorry about my poor english.
        Reply
        • Ludo Waltman April 17th, 2018 2:46 pm
          The network structure offers the most detailed information about the way in which keywords relate to each other. The clustering provides a simplified picture by assigning each keyword to a single cluster. Indeed this means that sometimes different items are assigned to different clusters even though the items are linked to each other. The clustering technique used by VOSviewer tries to assign items to clusters in such a way that the number of links between items that are in different clusters is minimized, but there will always be some links that connect items in different clusters.
          Reply
  • Tanya March 26th, 2018 5:24 pm
    Hi VOSviewer (and thanks for the wonderful support you provide here!),

    I'm trying to launch VOSviewer through the command line to increase available memory. I've downloaded the .jar file and run the code 'java -Xmx1000m -jar VOSviewer.jar' in terminal, however before VOSviewer launches I get a very long error, starting with:

    Exception in thread "AWT-EventQueue-0" java.lang.NullPointerException
    at org.pushingpixels.substance.internal.utils.SubstanceColorUtilities.getDefaultBackgroundColor(SubstanceColorUtilities.java:759)

    I'm not sure how to proceed from here..? Thanks!
    Reply
    • Ludo Waltman March 26th, 2018 6:22 pm
      Most likely, this error is caused by the fact that you are using Java version 9. Unfortunately, VOSviewer currently cannot be run using Java version 9. You need to use version 8 of Java. In the future, we of course hope to make VOSviewer compatible with Java version 9, but this requires some significant technical changes to be made to the software.
      Reply
      • Tanya March 27th, 2018 4:32 pm
        Ah, thanks! I had installed JDK10 in order to run the java command in terminal. I switched to JDK8 and was able to expand the memory with no problems.
        Reply
  • Antonio February 26th, 2018 2:27 pm
    Hello Ludo.
    I've just updated to new version 6.7, but now I cannot generate the network of authors according to their bibliographical coupling. The option "author" is not avaible anymore, but "documents", "sources", "organizations" and "countries" are still available. When I try to identify the authors by their "documents", all them appear as "anonymous".
    I tested with the same .txt files extracted from Web of Science in the 6.6 version and it worked well.
    Thanks for this powerful software!
    Reply
    • Antonio February 26th, 2018 2:42 pm
      Just a correction: this problem occurs when I export the results from SciELO Citation Index (a Web of Science collection). But the same .txt file works well with 6.6 version.
      Thanks again!
      Reply
      • Ludo Waltman February 26th, 2018 3:40 pm
        Could you please send your file to vosviewer@cwts.leidenuniv.nl? I will then take a look at it. Thanks!
        Reply
        • Ludo Waltman February 26th, 2018 11:12 pm
          Thanks for sending your file. In VOSviewer 1.6.7, we changed the way in which author data is extracted from Web of Science files. This was done in order to be able to process first names of authors. The new functionality was tested for data downloaded from the Web of Science Core Collection. We did not perform any tests for the Web of Science SciELO citation index. VOSviewer 1.6.7 indeed turns out to be unable to extract author data from Web of Science SciELO files. We were not aware of this, and we will try to fix this problem in the next update of VOSviewer. As a temporary workaround, you could use VOSviewer 1.6.6 for creating maps of authors based on Web of Science SciELO files. Alternatively, in your files, you could change the header 'AU' into 'AF'. By changing the header, VOSviewer 1.6.7 should be able to process your files. Thanks a lot for drawing our attention to this issue.
          Reply
  • Sven Januszek February 22nd, 2018 2:44 pm
    It seems as if the bug from Scopus has not been fixed yet. I am still facing problems using my CSV files from Scopus. It says: Error while reading Scopus file "XXX" (line: 5974): Incorrect number of coumns.
    Reply
    • Kirsty February 22nd, 2018 2:51 pm
      Yes it is unfortunately - apparently Scopus have said it should be fixed by the end of the month...
      Reply
    • Ludo Waltman February 22nd, 2018 5:02 pm
      Indeed the problem has not yet been fixed. The Elsevier Scopus team is still working on it. They expect the problem to be fixed by the end of the month.
      Reply
      • Ludo Waltman February 23rd, 2018 9:31 am
        I have just been informed by the Elsevier Scopus team that an update of Scopus will take place on Wednesday February 28. This means that on Thursday March 1 the problems with the Scopus CSV export should be solved.
        Reply
        • Ludo Waltman March 1st, 2018 10:50 am
          I have just performed some tests with the Scopus CSV export. It seems that the problems have been fixed. If anyone is still experiencing problems, then please let me know.
          Reply
          • Sven March 2nd, 2018 9:28 am
            Seems to still not work for me. I exported the CSV file from Scopus this morning and still get the error message: "Error while reading Scopus file 'XXX' (line 9711): Incorrect number of columns.
            Maybe my excel sheet is just too large with almost 10k rows.
            Reply
            • Ludo Waltman March 2nd, 2018 9:43 am
              Could you please send your file to me (vosviewer@cwts.leidenuniv.nl)? Thanks!
              Reply
              • Sven March 2nd, 2018 9:59 am
                I sent you the files and hope the mail attachment was not too big. Please let me know if you still have not received the mail.
                Reply
                • Ludo Waltman March 2nd, 2018 4:43 pm
                  Thanks Sven for sending me your files. I have taken a look at them.
                  Scopus has two options for exporting CSV files, an option ‘CSV - Only the first 2,000 documents’ and an option ‘CSV - Export all documents to a CSV file, citation information only’. It seems that you have used the second option, which apparently still suffers from a bug. I will report this to Elsevier. However, my recommendation is to use the first option. Your CSV file will then include more data per document, so that you have more options in VOSviewer for creating maps. It seems that the first option does not suffer from bugs anymore. If you experience any problems with this option, then please let me know.
                  Reply
                  • Sven March 5th, 2018 9:01 am
                    Thanks a lot Ludo!
                    Indeed the first option works out for me. Unfortunately, I would have to use more than just 2000 entries for my analysis. Hopefully, the second option will get fixed soon by Scopus. In case you hear anything from them, it would be nice to get an update from you.
                    Cheers,
                    Sven
                    Reply
                  • john March 13th, 2018 11:22 am
                    Hi,
                    Unfortunately, I'm still not able to generate a map in VOS Viewer based on a Scopus CSV output file. The program reports: Incorrect number of columns. Is this somehow related to the previous reported bug? Or maybe I made a different mistake?
                    Kind regards,
                    John
                    Reply
                    • Ludo Waltman March 13th, 2018 12:36 pm
                      I am sorry to hear this, John. Could you please send your file(s) to vosviewer@cwts.leidenuniv.nl? I will then take a look at them.
                      Reply
  • Christian February 18th, 2018 12:44 pm
    How are the total link strengths calculated?
    Reply
    • Ludo Waltman February 18th, 2018 1:50 pm
      This is explained in the VOSviewer manual. If the explanation provided in the manual is insufficient, then please let me know.
      Reply
      • Peter Frappell February 21st, 2018 7:30 am
        Hello Ludo -
        I get the "Links" attribute it is directly related to the number of co-authorship linkages, but I am also unsure how you weight the "Total Link Strength" attribute, could you please explain.
        thanks, Peter
        Reply
        • Ludo Waltman February 21st, 2018 6:12 pm
          Let me try to clarify this by providing a simple example. Consider an author A. Suppose author A has 4 co-authored publications with author B and 5 co-authored publications with author C. Author A then has co-authorship links with two other authors, and therefore the total number of links of author A equals 2. The total strength of these links equals 4 + 5 = 9. I hope this clarifies the distinction between 'links' and 'total link strength'. Please let me know if it is not yet sufficiently clear.
          Reply
          • Peter February 21st, 2018 10:36 pm
            This simple example makes it perfectly clear, Perhaps the definition in the manual could be reworded:
            “Links” - number of co-authors
            “Total link strength” - ithe sum of the number of links with each co-author.
            Thanks for the prompt reply.
            Reply
  • Stephane ledoux February 14th, 2018 11:39 am
    i've been trying to process a larger than usual set of data and failed, i've try to increase the memory allocation by using the command line java -Xss1000k -jar VOSviewer.jar after i've downoaded the Vosviewer.jar file it has indeed open vosviewer but the allocation was stuck to 247mb it is even lower than when i used vosviewer 1.6.6 directly (989 mb)
    Reply
    • Ludo Waltman February 15th, 2018 4:53 pm
      You need to use -Xmx instead of -Xss.
      Reply
  • Ka Pauw January 24th, 2018 3:32 pm
    Hi,
    How do you generate different variations of the term map with regard to journals, countries and open access?
    I have noticed this feature in your presentation "VOSviewer: A software tool for analyzing and visualizing scientific literature" [Sep 14, 2017] page 11-13 on SlideShare, but when I "Create a map based on text data" based on Web of Science files I do not see any options in the interface to associate the coloring with any underlying data attributes.
    Thanks for a very practical and yet powerful piece of software!
    Reply
    • Ludo Waltman January 24th, 2018 5:33 pm
      When creating a term map based on Web of Science files, these options are indeed not available. Instead, you need to create a term map based on a VOSviewer corpus file and a VOSviewer scores file. This means that you need to do some preprocessing of your data. You need to concatenate the titles and abstracts of the publications in your Web of Science files and you need to save the concatenated titles and abstracts in a corpus file. In addition, you need to create a scores file, with scores indicating whether a publication has appeared in a specific journal, whether it has been authored by a specific country, or whether it is open access. Creating the scores file can be a bit more complex. More information about scores files is provided in the VOSviewer manual.
      Reply
  • Sri Amudha Kumaresan January 13th, 2018 3:47 am
    Hello! Thank you so much for sharing this wonderful software. I'm using this software in my research for cluster analysis. I have done my work in version 1.6.4 and due to technical difficulties I have lost it. Now, I'm unable to download the same version. Can you please share me the version 1.6.4?
    Reply
    • Nees Jan van Eck January 15th, 2018 9:46 am
      VOSviewer version 1.6.4 can be downloaded using the following links:
      MS Windows: http://www.vosviewer.com/download/f-x2z2.zip
      Mac OS X: http://www.vosviewer.com/download/f-y2r2.zip
      Other systems: http://www.vosviewer.com/download/f-y2q2.zip
      Reply
  • Theoni Paschou January 1st, 2018 5:23 pm
    Good afternoon and a happy new year,
    I saw your software applied to another literature review (Zhang, 2017) and I am also interested in it but I face some difficulties importing the data.
    Can I import them also from excel? I have the Impact factor, number of citations, year etc.
    Otherwise would you mind supporting me so to import the data from Scopus?
    I would truly appreciate your kind help.
    Best,
    Theoni
    Reply
    • Ludo Waltman January 2nd, 2018 11:45 am
      If you want to provide your own data as input to VOSviewer, this can be done using VOSviewer map and network files, which can be created using Excel. The format of these files is documented in the VOSviewer manual. When you download VOSviewer, an example of a map and a network file is included in the ZIP file that you will obtain. However, creating your own map and network files is something which requires an in-depth understanding of VOSviewer. If you have only limited experience with the software, I don't recommend to take this approach.
      If you want to use Scopus data, you need to download data using the CSV export feature in the Scopus web interface. The CSV files obtained in this way can be provided as input in the 'Create Map' wizard in VOSviewer. However, at the moment, there is a bug in the Scopus CSV export, and therefore this approach may not work. Elsevier is working on fixing the problem. An alternative could be to work with Web of Science data.
      Reply
  • Ali H December 29th, 2017 2:06 pm
    Dear Friends
    When I open a CSV from scopus, most time I see error:
    Incorrect number of columns
    It's very time consuming to repair it by finding the incorrect columns. can you help?
    Reply
    • Ludo Waltman December 29th, 2017 2:29 pm
      Unfortunately, there currently is a bug in the CSV export of Scopus. This is causing the problem that you mention. We are in contact about this issue with the Scopus team. They are working on fixing the problem. I hope the problem will be solved soon.
      Reply
      • Kirsty S February 12th, 2018 3:22 pm
        Does anyone have a Scopus file that works that they might be able to share (or at least share the column headers) so that I can work out which columns are required?
        Reply
        • Ludo Waltman February 12th, 2018 3:30 pm
          Please send me an e-mail (waltmanlr@cwts.leidenuniv.nl) and I will try to help you with this.
          Reply
  • Marcela C December 19th, 2017 4:08 pm
    Hi,
    I'm using VOSviewer 1.6.6 for a bibliometric study based on records from Web of Science.
    When performing "bibliographic coupling" on documents, why do I still get results in terms of "total link strength"? Shouldn't I get a different score such as "bib. coupling"?
    Thank you
    Reply
    • Ludo Waltman December 19th, 2017 10:09 pm
      In your case, 'total link strength' can be understood as the total bibliographic coupling strength. VOSviewer currently does not adjust the terminology it uses to the specific type of analysis that is carried out. In future versions of the software, we do plan to adjust the terminology used by the software to the type of analysis.
      Reply
  • dennis December 4th, 2017 11:01 pm
    hello,
    is writing the thesaurus the only way to clean data? because when you have thousands of keywords, writing the thesaurus is really time consuming. Maybe there is another software to assist the cleaning phase? what do you suggest?
    Reply
    • Ludo Waltman December 4th, 2017 11:34 pm
      Yes, this is the only option provided by VOSviewer for cleaning your data. I don’t have specific suggestions for other software tools that you could use for this purpose. However, I do want to mention that in many cases superficial cleaning of keywords may actually be sufficient. As developer of VOSviewer, my recommendation usually is not to focus too strongly on individual keywords appearing in a map, but instead to take a more holistic perspective focusing on the overall patterns emerging from a map. If the latter approach is taken, detailed cleaning of keywords usually is not necessary, and more superficial cleaning is likely to be sufficient. Hence, I often recommend to try to be pragmatic in data cleaning.
      Reply
    • Gilberto Parra January 28th, 2018 9:01 am
      You can try OpenRefine
      Reply
  • Evan Sterling November 30th, 2017 5:25 pm
    The Organization unit option does not appear when I create a map from Scopus data, even though I have double-checked that the CSV files include the Affiliation column, and have tried with multiple data sets. Can you suggest any reasons this might be happening? My export includes all available Scopus fields.
    Reply
    • Ludo Waltman November 30th, 2017 7:30 pm
      The option to create co-authorship maps of organizations based on Scopus data is no longer supported by VOSviewer. This option has been removed in the most recent version of the software (version 1.6.6). The reason is that Scopus (unlike Web of Science) doesn't perform any harmonization of organization names. We therefore felt that Scopus data is not sufficiently clean to create proper co-authorship maps of organizations.
      Reply
  • Dlan November 8th, 2017 1:36 pm
    Hi Ludo,
    When running VOSviewer, I encountered a problem pertaining to being out of memory.
    Number of articles - 35,100
    Numbers of input txt files - 71 (web of science)
    Based on the VOSviewer Manual, I attempted to increase the memory for VOSviewer. However, I could not find the VOSviewer JAR file.
    I downloaded the zipped file of VOSviewer and unzipped it in a different folder. From this folder, I ran the exe file of VOSviewer. In this folder and in my computer, there is no VOSviewer JAR file.
    My computer has 8 GB RAM.
    I would be glad if you could help solve my problem.
    Thank you.
    Reply
    • Nees Jan van Eck November 9th, 2017 2:18 pm
      The VOSviewer JAR file can be obtained using the link 'Download VOSviewer 1.6.6 for other systems' on the Download webpage.
      Reply
  • Daniel October 22nd, 2017 2:42 pm
    Awesome software! Thank you for making it available to all! Is it possible to merge two databases, one extracted from Web of Science and the other from Scopus and have the data analyzed it jointly by VOSviewer?
    Reply
    • Ludo Waltman October 23rd, 2017 10:34 pm
      Please see below the question by Ana on September 2nd, 2015.
      Reply
  • Charis Papaevangelou October 22nd, 2017 12:59 pm
    Hello!
    First off, I'd like to congratulate and thank you for your amazing work!
    I have two quick questions to ask:
    1) Will it be possible in the future to change clusters/items in real-time? For example, while working with one corpus, and you decide that it's actually not that meaningful and want to remove it.
    2) Are other languages going to be supported someday? I'm interested in working with Greek, which is my first language, and VOSviewer does not currently support them, to my understanding.
    Thank you once again :)
    Reply
    • Ludo Waltman October 23rd, 2017 10:32 pm
      Thanks Charis for your suggestions. Although we do not yet have concrete plans for the two points that you mention, we will keep your suggestions in mind for future versions of VOSviewer.
      Reply
  • Pedro October 11th, 2017 2:23 pm
    Topic: Bibliometric data analisys based on keyword occurence.
    Is it right to say that the lines connecting the keywords on my network are a representation / description of how many time those keywords were used simultaneously?
    Could I infer more information from this lines interpretation?
    Cheers and thanks beforehand!
    Reply
    • Ludo Waltman October 11th, 2017 5:56 pm
      Yes, that is correct, the strength of a line (link) indicates the number of documents in which two keywords are used together. The thickness of lines in the visualization is proportional with their strength, although it may be difficult to visually distinguish between lines with different thickness.
      Reply
      • Mihaela October 15th, 2017 5:12 pm
        Regarding the difficulty to distinguish between lines. Is there any way to export these data and import into another program (excel for example) in order to sort by link strength and identify the strongest links?
        Reply
        • Ludo Waltman October 15th, 2017 10:28 pm
          Yes, this can be done by saving a network in a VOSviewer network file. The 'Save' button on the 'File' tab in the action panel of VOSviewer can be used for this purpose. A VOSviewer network file is a simple plain text file that can be imported for instance into MS Excel.
          Reply
  • Isabel Pinho October 2nd, 2017 12:06 pm
    In this article:
    Van Eck, N.J., & Waltman, L. (2017). Citation-based clustering of publications using CitNetExplorer and VOSviewer. Scientometrics
    In pag 1060, fig1-show 22 clusters from CiteNetExplorer
    Next, in pg 1063, fig 3- using VOSviewer to analize those 22 clusters
    How Can I do the same?
    If you please, can you give the tutorial steps to make the visualization in VOSviewer of the 22 clusters that can from CitNet Explorer?
    Reply
    • Ludo Waltman October 3rd, 2017 12:03 am
      Performing an analysis similar to what is reported in this paper is not entirely trivial. It requires an extensive knowledge of VOSviewer and CitNetExplorer and also some data processing skills. As we write on p. 1065 in the paper, “the combined use of the two software tools is somewhat laborious, and preparing the input data for VOSviewer based on the clustering results provided by CitNetExplorer is not entirely straightforward”.
      Basically, you need to take the following steps:
      1. In CitNetExplorer, cluster the publications and export the clustering results. To export the clustering results, choose the option ‘Export groups – Full network’ in the ‘Groups’ menu. Also make sure to save the citation network in a publications and a citations file (using the ‘Save’ button).
      2. Based on the cluster assignments of publications as well as the citation links between publications, it is now possible to construct a citation network of clusters. This requires some data processing skills. Constructing the citation network of clusters for instance can be done using MATLAB, R, or Python.
      3. The citation network constructed in step 2 needs to be saved in a VOSviewer network file. In addition, a VOSviewer map file needs to be created, providing basic information for each of the clusters (e.g., a label and a weight, where the weight can be calculated as the number of publications belonging to the cluster). More information about map and network files is provided in the VOSviewer manual.
      4. In VOSviewer, click ‘Create’, choose ‘Create a map based on a network’, select the map and network files created in step 3, and follow the remaining steps in the ‘Create Map’ wizard.
      I hope the above explanation is helpful, but I understand that some of the steps, especially step 2, may still be quite challenging.
      Reply
  • Minh Nguyen September 4th, 2017 4:23 pm
    Could you elaborate on how the relevance score is calculated in the creation of map based on text data?
    Reply
  • James Qin September 1st, 2017 7:05 pm
    Is it possible to restrict the analysis to a subset of items in VOSviewer published in a specific period? If yes please let me know how to do that. Thanks!
    Reply
    • Ludo Waltman September 2nd, 2017 10:37 am
      This currently is not possible. It is a feature that we hope to add in some future version of VOSviewer.
      Reply
  • Khan jee August 22nd, 2017 9:03 am
    It's a nice endeavor, i appreciate as is very much helpful in research domain.
    Reply
Build on Applepie CMS by Waltman Development