I just noticed that Research Gate does not detect References in Book Chapters and Conference Papers nor Posters, even though these have an explicit References section, thus, they do not add citations to those cited in them. Why is that?
The only way to try to keep the RG citations metric up-to-date is to ask citing you authors for uploading a given paper that cites your works as full text. Even if a given paper has copyright limitations you can ask them to upload it as a private copy, so it won't be visible by other but lets RG platform to find citation of your work in it.
Anna Małgorzata Kamińska What if RG does not have your work listed (you cannot see the References displayed), nor the work citing your work, either? I thought it's a matter of time before it's updated, but not really
1. If someone uploaded his pdf file containing an article with references, and some of these references is citing your work, you cannot verify it if the upload was as private, because it is visible only to the person who has uploaded it
2. Citations are calculated "automagically" by parsing references sections of uploaded articles. Unfortunately quite often this parsing is failing (because of a wrong structure of pdf, special characters in references records etc. ) what leads to underestimation of citations counts. And unfortunately because of point 1. we even cannot find the reason.
Anna Małgorzata Kamińska The "automagical" update is what concerns me. For instance, it does not detect all the references in my theses, and no references at all in other book chapters and conference proceedings. I uploaded my own copy for the theses, and the official final copy I got from the editor(s) for the other published documents.
Research Gate should allow us include/upload references just as we upload figures, and improve their system. Otherwise, this factor is definitely not reliable as a reference. For now, I've noticed that they update bit a bit, and the references count does change in the theses case (although it's far from accurate).
you are right indeed. It is enough to look at example:
Article The application of methods of social network analysis in bib...
The system has detected 8 reference records while in the full text we see 13.
But letting authors for manual adding ones could expose the system for fraud issues. Of course, presently authors also can prepare and upload fake pdf with fictitious references, so here there is not a perfect solution unless it will be controlled at some stage of data collecting. Also, the question remains here what exactly a scientific work is and should RG platform calculate only scientific references or all of them, etc.
Meanwhile, I propose not to pay to much attention at RG indicators and not treat it as a bibliographic database, but as service helping in scientific cooperation mostly on a social ground :) what exactly we do writing these posts ;)
But regarding your proposition, manually adding reference records still doesn't solve problems of various citation styles. For strict calculation of citations, the system should uniquely identify each matching reference record. But matching here should be understood as a logical match, not a physical because many different character strings can refer the same bibliographic unit. So things are not as simple as may appear.
Anna Małgorzata Kamińska I've seen researchers use their RG index when writing porposals and attaching their CV, as a quality reference. I am not as familiar with RG, as I've mostly used it as an enriching source of information, hence, I was surpirised the index calculated here could actually be used for anything at all.
As for what exactly a scientific work is and whether RG platform should calculate only scientific references or all of them... that's a very broad topic. I've worked as a journal manager and we did not accept citation of posters, BSc theses, abstracts, or sometimes even MSc theses, unless they underwent a thorough examination (similar to published articles). And yet, it was obvious that sometimes References were prepared in such a poor way that they were hardly useful for the reader unless corrected. That gave me the hint that plenty of researchers do not give the deserved (nor understand the) value to (of) References.
Subir Bandyopadhyay Christopher Nock What about Google Scholar? It's search and identification tool is not fully functional, as I see it. It identifies some of my publications with the wrong title and authorship (which I have not managed to fix so far) and it detects the wrong source document for others (I have not managed to solve this problem either, even though I've deleted and reuploaded the original files)
I only refer to Google scholar. It's better than RG for publications but not perfect. There are ways of contacting Google about problems via email. My wife knows how, I don't!
I am rather familiar with using indexed journals as a reference, and "measuring" author's activity based on where they publish, the quality, how active they are and how fruitful their contribution is to my work... I have not found a single index that is perfect so far... they all seem rather biased or partially lacking. Christopher Nock Could you ask her, please? Otherwise I will look it up myself, I am sure I can find something if I dig enough
She tells me she can't remember! It generally involves finding a help button and keep entering things it can't answer until it offers the option to email.....
I completely agree with Dr. Arvind Jayant. I have seen many researchers having high RG score but that is mostly contributed by their participation in discussion and more importantly how much they recommend others' works even those do not belong to their area of research!
I completely agree with all respondents. Adina Moraru Arvind Jayant José Carlos Sá Sanjib Biswas Christopher Nock Subir Bandyopadhyay
RG must explain how they calculate the score. If you want to get a high score, you can easily do that by answering questions rather than uploading the book chapter.
GS is one click better than nothing and WoS and Scopus are subscription databases with higher academic reputation. There is a lot of material in GS, but this does not indicate quality, since much of included content cannot be classified according to type of article, research interest, country of origin etc.
Because that citations depend only on RG database, I think the only solution is to ask RG members to upload their articles as private copy. In this case, RG can add the citations without violate publication copyright.
RG citations calculation is an online process. It is measure of knowledge of an expert useful for other community . It depends up on quality of the answer and quality of the paper having good impact factor.