I would recommend to ask this Q at Stack Overflow community too:

https://stackoverflow.com/

Hi Arthur

The column DB should have provided the information you are looking for, however, it seems after merging it has a default value of ISI (WoS). However, the column UT contains the IDs using which you can distinguish between Scopus and WoS data. The WoS IDs start with ISI (for example, ISI000165584400003) whereas the Scopus IDs are a bit random. Just have a look and you can easily figure it out. The only issue is that it will be hard to identify the intersection part as I think the common ones represent either of the IDs (but not both). If you need to keep track of all the information such as the Scopus and WoS records and the common records, then you might have to customize your code a bit. You can start by setting the argument remove.duplicated in the mergeDBSources function from the bibliometrix package as FALSE. This will keep all the records including the common ones between Scopus and WoS. Thanks.

What we have detected is that in Scopus the surnames and names that have accents and characters in Spanish these continue to appear and those of the WoS everything is without accents. We also see that in WoS there are many poorly accented Latin authors and they are indexed with their second surname, also confining names with surnames. Likewise, I recommend that you use the desktop version of EndNote when downloading the records from both systems, which will allow you to identify from which source the record is and at the same time eliminate duplicate records between both systems.

Glad to help. You may have to be a bit careful around merging though. It looks like a good but temporary solution and may have issues with some data. I had come across duplicate records in individual groups such as Scopus or WoS with different UT values. Even the merging by the function mergeDBSources is not always up to the mark. The basic working of this function is matching the TI column which is the title. I am not very sure but sometimes even for the same article, different titles are recorded or the same title with varying values for other columns. Got to think that is there actually a possibility of more than one article having the same titles? In addition to that many of the times, some of the columns have NA values which make the merging really difficult. The function mergeDBSources ignores this factor and merges.

As I have already mentioned custom scripts are a better idea, however, the data used plays the most important role in merging. If the data, you are using has no duplicated records from a single source and no missing value, then it should work fine. Another suggestion would be to work around PubMed IDs to get the unique records, however, the fact that all WoS or Scopus articles are not in PubMed comes with a real obstacle.

I still have no idea why all the value for the column DB after merging has forced to be ISI. Cheers.

You may take a look at this video:

https://www.youtube.com/watch?v=BxrKYfPZ59o

N

Hello. This article may be really useful for your issue Article A user-friendly method to merge Scopus and Web of Science da...

Article A user-friendly method to merge Scopus and Web of Science da...