Bulk access to data on scientific literature (affiliation, citations etc)

Question

I want to compare the scientific performance of different countries and/or institution via the relation of first and second etc authorship of papers in relation to the number of citations of each paper to answer specific questions.

Therefore, i need a dataset that covers the following aspects of publications:

• All author's affiliations with categorial indexing (1st or 2nd etc), including the address of the institution (just the country actually)
• Number of citations
• (Title and/or) field of study
• For a representative number of papers in as many fields as possible (e.g. Earth Sciences, Medicine, Math etc)

So the question is: Where do i get this kind of data in bulk, preferably free?

What i have tried so far:

• Web of Science, which can cover part of the information i'm looking for, but basically blocks downloading it.
• A basic crawler that scrapes the info from an individual journal's web page. However, i don't have the computing power and coding experience in this field for this to be time efficient.

Answer 1

I am not sure that you can get everything you are looking for directly, but here are a few sites that have some of the information you're looking for:

Cite Seer: http://citeseerx.ist.psu.edu/

Social Science Research Network: http://www.ssrn.com/

Google Scholar: https://scholar.google.com/

Note that the 'free' and 'open' -ness of these sites should be researched before you go too far.

A couple of other sites that might be relevant, but less open:

ResearchGate: http://www.researchgate.net/

Science Direct: http://www.sciencedirect.com/

Good luck!

Answer 2

RePEc

• https://ideas.repec.org/getdata.html

Although for papers in economics alone, RePEc has several advantages for your project:

• Data is free: http://repec.org/docs/RePEcDataUse.html
• Full FTP access to the structured data
• Multiple metrics for authors
• Broad data coverage

Answer 3

For anyone still looking for sources of citation network data, and paper metadata for global cross-disciplinary publications, the Open Academic Graph project from Microsoft and AMiner combines two of the largest indexes, and is free.

https://www.openacademic.ai/oag/

If you are interested in downloading full content (i.e. PDF/LaTeX source), then arXiv will allow you to do this for about $25 in data transfer costs. This is for Computer Science papers:

https://arxiv.org/help/bulk_data_s3