ISI Web of Science/Knowledge APIs March 24, 2009
Posted by jrochkind in General.trackback
While you won’t find much of a clue in your google searching, ISI does offer some limited APIs and direct-linking-syntaxes for the Web of X products.
(More info on my subsequent use of ISI and Scopus webservices can be found in this post.)
It seems you can use many of them with an existing ISI subscription, no additonal charge.
You do need to register/request access.
Here’s one that looks particularly useful that I plan to use in Umlaut:
http://isiwebofknowledge.com/products_tools/products/related/trlinks/
If it works as I have been told, it will allow me to look up a particular citation in ISI, and get links to the ‘cited by’ page for that citation, ISI-computed ’similar articles’ for that citation, and perhaps others.
I don’t want to display any of these links to my users unless I know clicking on them is going to give non-zero hits. Hopefully the API will tell me that, otherwise my app will just have to do a request in the background first and screen-scrape the HTML to make sure there are non-zero hits.
Either way, it’s going to be a LOT easier than trying to simulate user interaction with ISI’s actual end-user web site, requiring lots of session cookies and lots of slowness.
Any idea where I can find info on available APIs for Web of X products? And how do I register/request access?
Cheers
Nick
Hi Nick, the URL to the registration form for the API I am using is in the post above; after you register you’ll get documentation emailed to you. That’s the only API I know about — the “Link Article Match Retrieval Service” is what I’m using.
If that doesn’t work for you, email me and I can give you my internal Thomson contact’s email, but I believe it will.
[...] ISI Web of Science/Knowledge APIs « Bibliographic Wilderness (tags: ISI api webservice) b92 [...]
Hi,
This seems very promising indeed. I’m actually interested in retrieving ‘times cited’ for a list of authors in an automated way. This API could come in handy. I’ve checked the link you provided here. Then I have to enter an IP-address in the form. The text next to that says:
“Please enter only the IP address of the institutional contact who will entitle the service for your site. Please do not enter an IP range. Once this institutional contact is entitled for use of article match retrieval service, they will not have access to the institution’s subscription to ISI Web of Knowledge.”
Does the last sentence actually mean that I will have to use either the API or the regular webaccess? Or can I use both at the same time? Does this in any way affect our institution’s end-users way of accessing the WoS? I’d just like to do some tests first.
Kind regards,
Richard
Richard, I recommend you contact ISI tech support for more info.
But my understanding is that, basically due to a weird unintended ‘bug’ on the ISI side, the particular IP address you enter to receive access to the API, will not have access to the web interface, yeah. Your institution can still have access to the web interface, just not that IP.
Normally this isn’t a problem, because that IP is a server that a web user can’t be coming from anyway. It can be a problem if your institution uses some kind of NAT/PAT such that your server appears to be coming from the same IP address as actual workstations with web browsers.