OpenAlex just released its API and its webiste is coming soon, which is exciting.
Search by DOI #
If you want to get data from OpenAlex, but don’t have openalex work id, you can get the same information via paper DOI. The API Dcoumentataion mentions this .
For example, you have this DOI: 10.1109/TVCG.2020.3030380.
First, you can access this paper, via two urls:
- http://dx.doi.org/10.1109/TVCG.2020.3030380
- https://doi.org/10.1109/TVCG.2020.3030380
These two urls point to the same destination: https://ieeexplore.ieee.org/document/9222338.
Okay, now, tell me how to access OpenAlex data based on DOI.
According to the The API Dcoumentataion , you can access it in these ways:
- https://api.openalex.org/works/doi:10.1109/TVCG.2020.3030380
- https://api.openalex.org/works/http://dx.doi.org/10.1109/TVCG.2020.3030380
- https://api.openalex.org/works/https://doi.org/10.1109/TVCG.2020.3030380
Search by title #
According to official documentation , we can search by title this way:
https://api.openalex.org/works?filter=title.search:cubist
The above url will show all the papers whose title contain the word of cubist.
If you have the exact title, just put it at the end of the url, for example,
https://api.openalex.org/works?filter=title.search:Efficient simplification of point-sampled surfaces
If the title your are quering contains ‘&’, ‘:’, ‘?, and/or ‘,’, rememeber to replace them with ''. This is because both ‘&’ and ‘:’ are quering parameters and will rusult in errors when quering.
If you use python, you can do it this way:
import re
title = 'this: is a bad title & not recommended for you, right?'
title_converted = re.sub(r'\:|\?|\&|\,', '', title)
Search for host venues #
Reference:
- https://docs.openalex.org/api/get-lists-of-entities#venues-filters
- https://docs.openalex.org/about-the-data/venue
By display name #
https://api.openalex.org/venues?filter=display_name.search:communication and sport
By ISSN #
https://api.openalex.org/venues?filter=issn:2374-3670
By ISSN_L #
https://api.openalex.org/venues/issn_l:2167-8359
Note that if you use ?filter=, then the results are usually more than one.
Pagination #
In list of entities, OpenAlex’s default is 25 items per page. You can view the first 10000 items by setting page and per-page parameters.
For example,
https://api.openalex.org/concepts
You can set them this way:
https://api.openalex.org/concepts?per-page=50&page=50
Another example,
https://api.openalex.org/works?filter=host_venue.id:V84775595
You can set the two parameters this way:
https://api.openalex.org/works?filter=host_venue.id:V84775595&page=2&per-page=50
I learned the trick of & here: Everything You Need to Know About API Pagination
by J Simpson
.
Query speed #
- Quering paper title is much faster than quering OpenAlex Work ID, or paper DOI.
- Being in the polite pool (i.e., disclosing your email) is not faster than in the common pool.
- The speed of two query methods for DOI, i.e.,
works/doi:DOIandworks/https://doi.org/DOI, do not vary significantly. Each query roughly takes 6-10 seconds. - The speed of quering OpenAlex Work ID and that of querying paper DOI do not differ significantly. Querying DOI might be slightly faster.
Last modified on 2022-04-11