Getting Started

At the moment, CitNetExplorer does not support Scopus files and Pajek files. I am afraid we currently do not have the resources to update CitNetExplorer to provide support for these file types. Hopefully at some point in the future such an update can be made.

This probably relates to the issue of early access publications discussed below (see my response to the question of Fernando Alguacil). My recommendation is to exclude early access publications from the data that you download from Web of Science. If this doesn't solve your problem, then feel free to send your files to citnetexplorer@cwts.leidenuniv.nl. I will then take a closer look at your problem.

There is a new feature in Web of Science, so-called early access publications. If your data includes such early access publications, CitNetExplorer will not be able to process the data and will give an error message. This probably explains your problem. My recommendation is to exclude early access publications from the data that you download from Web of Science.

Could you send your files to citnetexplorer@cwts.leidenuniv.nl? We'll then try to take a look at your problem.

Yes, in the long term, we hope the software will evolve in such a direction. In the short term, however, I'm afraid we don't have the resources to work on this.

Your question seems to be more about CRExplorer than about CitNetExplorer. I'm afraid I'm unable to answer your question.

Could you send your first file to citnetexplorer@cwts.leidenuniv.nl? We will then take a look at your problem.

I am afraid that CitNetExplorer doesn't provide any support for identifying the language of a publication.

I don't fully understand your question. What exactly do you mean by 'citation network patterns'? Perhaps the CitNetExplorer tool is more suitable for studying such patterns?

I have very limited practical experience with CRExplorer, but I believe this can be done using the import and export features available in the 'File' menu.

You need to download data from Web of Science not only for your publication of interest but also for all publications citing this publication. VOSviewer is less suitable for showing the citation impact of one specific publication. Like CitNetExplorer, VOSviewer can be used to visualize citation networks of publications. However, the visualizations provided by CitNetExplorer will probably be more useful than the ones provided by VOSviewer.

No, in CitNetExplorer there is no support for thesaurus files. Individual publications, not authors, are the unit of analysis in CitNetExplorer. Therefore I don't think there is a strong need for thesaurus files in CitNetExplorer.

Yes, up to 100 publications can be visualized. Use the 'Update publications' button to increase the number of publications in the visualization.

Could you please send one of your files to citnetexplorer@cwts.leidenuniv.nl? I will then take a look at your problem.

It might be that this problem is due to the use of a 32-bit version of Java. Make sure you are using a 64-bit version of Java. If this doesn't solve the problem, then please let me know.

I assume you are working on a Mac? This is a known issue when CitNetExplorer is used on Mac systems. Unfortunately, I don't have an easy solution. My recommendation is to use a Windows system to make screenshots. In the future, we hope to fix the problem, so that screenshots can also be made on Mac systems.

You are completely right. Due to a bug in the software, the external citation scores don't work properly. It's best not to use this option. I hope you will find CitNetExplorer useful even without being able to use the external citation scores. Of course, when the software is updated, we will fix the bug.

Unfortunately, this is not possible. Google Scholar doesn't provide functionality for downloading citation data.

My suggestion is to first open a Web of Science file in CitNetExplorer and to then save the resulting citation network in a publications and a citations file. The resulting publications and citations files can be used to get a better understanding of the information that is stored in these two files. When you understand this, you may be able to create your own publications and citations files.

Please see my earlier response to a similar question asked by Gaurav Maheshwari.

I am afraid this will be very difficult. First of all, CitNetExplorer doesn’t perform co-citation analyses. (Instead, the VOSviewer software tool could be used for such analyses.) CitNetExplorer performs only direct citation analyses. Perhaps more importantly, processing PDF files is highly challenging and CitNetExplorer doesn’t offer any support for this. (The same applies to VOSviewer.) You need to take care yourself of converting your PDF files into text files, extracting the reference strings from the text files, and identifying the citation links between documents based on the reference strings. The data obtained in this way can then be provided as input to CitNetExplorer using so-called publication and citation files. However, it is clear that this is a highly challenging process that takes a lot of effort, so I can understand that this might not be what you have in mind.

No, networks are not constructed (exclusively) based on DOI. Other data elements are taken into account as well. For more information, please see Section 4.1 in the following paper: https://doi.org/10.1016/j.joi.2014.07.006 (http://arxiv.org/abs/1404.5322).
CitNetExplorer should work well on 64 bit Windows systems. There is no clear minimum amount of memory that is required by the software. The larger the amount of memory you have available, the larger the data sets that can be handled by the software. The 'file is empty' error message probably is caused by the problem discussed below in my response to the question of Alderasus.

Thanks for raising this issue. We have investigated it. There currently turns out to be a problem with the download feature of Web of Science. Because of this problem, files downloaded in the so-called tab-delimited file format cannot be opened by CitNetExplorer. They cause a 'file is empty' error message. We are in contact with Web of Science and they are trying to fix the problem. The problem should be fixed within the next one or two weeks. Most likely, this problem explains why you are getting an error message.
A workaround is to use the so-called Web of Science plain text file format. This format can also be handled by CitNetExplorer and it doesn't suffer from the above problem.

Unfortunately, CitNetExplorer doesn't yet offer Scopus support. However, there is a workaround. You can use another tool, CRExplorer (available at www.crexplorer.net), to open your Scopus data files and to save them as Web of Science data files. The Web of Science data files that you have obtained in this way can be opened by CitNetExplorer. In this indirect way, CitNetExplorer can work with Scopus data.

Assuming that you are working with data downloaded from Web of Science, you could edit the Web of Science files using a text editor. In that way, you can make changes to the references. However, this is quite complex and this needs to be done in an extremely accurate way, otherwise CitNetExplorer will no longer be able to process the Web of Science files.
Alternatively, after you have loaded a citation network in CitNetExplorer, you can save it using the 'Save' button. You will obtain two files, a publications file and a citations file. You can edit the publications file using a text editor (or using a spreadsheet program such as MS Excel) and you can then make changes to the data on individual publications.

Dear Alex,
I can still see these options on the web of science ISI. After selecting an article, click on Save to other file formats (dropdown bar on top of the article title) and then on the pop up window select Full record and cited references from Record Content and then tab delimited-Win on the file format for the extraction on a txt format.
Hope that helps, George

Alex, most likely you are not using the Web of Science Core Collection. When you go to the online Web of Science, by default the ‘All Databases’ option is selected. You need to change this into ‘Web of Science Core Collection’. I expect that this will solve your problem.

I looked at the way I imported the records from Web of Science and realized I had made a mistake.
I had left the citation data out.
I just did a full import of all the fields from WOS and now it works :)

Unfortunately there is no easy solution to this. OECD reports, Word Bank articles, and other publications that are not available in Web of Science cannot easily be included in a citation network that is analyzed using CitNetExplorer. If you want them to be included, you need to do your own processing of the references contained in these publications. You can then create your own publications and citations files and these files can be opened by CitNetExplorer. In this way, publications not available in Web of Science can be analyzed using CitNetExplorer, but it may take quite a lot of effort to preprocess the data in such a way that it can be used as input for the software. Please note that, if the ‘Include non-matching cited references’ option is turned on in CitNetExplorer, publications not available in Web of Science can be included in a citation network that is analyzed using CitNetExplorer if these publications occur sufficiently often (10 times by default) in the reference lists of other publications that are available in Web of Science.

Merging duplicate publications is not possible in CitNetExplorer. There is a solution, but this solution is somewhat technical. In CitNetExplorer, you can save a citation network into a publications and a citations file. These are simple text files that can be edited using a text editor or using a spreadsheet program such as Excel. In the publications file, each line represents a publication. If you want to merge two publications, you need to remove the line corresponding with one of the two publications. In the citations file, each line represents a citing link between a citing and a cited publication. After removing a publication from the publications file, you need to update the citations file in order to make sure that there are no citation links for which either the citing or the cited publication has been removed. Such citation links need to be updated. For instance, if the citing publication has been removed, it needs to be replaced by the corresponding duplicate publication.

1. In Web of Science, when saving bibliographic data into a file, only partial information about cited references will be available. For instance, only the first author of a cited reference will be listed. Also, the title of a journal article will not be listed. The journal name, on the other hand, will be included.
2. When you look at the list of the cited references of a publication in Web of Science, it is indeed not possible to save these cited references. In Web of Science, you can save bibliographic data on publications included in Web of Science. This will then also include some data on the cited references of these publications. Directly saving the cited references is not possible. Please note that, when CitNetExplorer constructs a citation network based on Web of Science files, cited references occurring in a sufficiently large number of publications (by default 10) will be included in the citation network (with incomplete data, so only first author and no title).

CitNetExplorer does not offer support for processing Scopus files. This is a feature that may be added in the future, although we have no concrete plans for this at the moment. Currently CitNetExplorer offers support only for processing Web of Science files. CitNetExplorer also has its own file format. This file format can be used to input Scopus data into CitNetExplorer, but you then need to do a lot of preprocessing yourself, which can be quite complex.

In the visualization of a citation network, CitNetExplorer has the restriction that publications can be labeled only using an author name, not using the title of the journal in which they have appeared. You can choose between the last name of either the first or the last author of a publication. When searching for publications on the 'Publications' tab in CitNetExplorer, it is possible to search based on journal title.

Yes, input files can be combined. You can simply download from Web of Science as many files as you need in order to cover all publications that you are interested in. All downloaded files can then be opened simultaneously in CitNetExplorer. In the file chooser dialog box in CitNetExplorer, you can select multiple files by clicking on them while pressing the CTRL key.

Thanks Tim for your feedback on CitNetExplorer. Did you select the 'Web of Science Core Collection' database? Usually, when people have problems downloading citation data, they are working in the 'All Databases' mode rather than the 'Web of Science Core Collection' mode. Using the 'All Databases' option, it is indeed not possible to download citation data.

I 'd really like to get this to work, too. You don't include any information about connecting to a citation database. I can only access WoS through my university's VPN in a browser. I have no idea how to tell your program how to get behind it. What am I supposed to do?
I can't imagine I am alone here with this. And what is with Google Scholar? That would be sort of obvious, wouldn't it.
In any case I succeeded in importing my bib from Citavi, (more or less), but I just can't move on from there.
Surely this is not supposed to be a solution only for people that are physically working on machines within university networks....?