If you have a product or service and need to know whether an email address belongs to an academic institution, there's a chance there's some manual component to the approval process or a you have a high false positive/negative rate. Perhaps .edu email addresses are automatically approved because, for the most part at least, they're associated with American post-secondary educational institutions. Perhaps .ac.uk email addresses are automatically approved because they're guaranteed to belong to British universities and colleges. Unfortunately, not every country has an education-specific TLD (Top Level Domain) and plenty of schools use .com or .net.
Swot is a community-driven or crowdsourced library for verifying that domain names and email addresses are tied to a legitimate university of college - more specifically, an academic institution providing higher education in tertiary, quaternary or any other kind of post-secondary education in any country in the world. Swot-bloom builds on this by creating a bloom filter from the Swot collection of institutional emails that can be serialised and easily used for client side verification.
Pop quiz: Which of the following domain names should be eligible for an academic discount? stanford.edu, america.edu, duep.edu, gla.ac.uk, wunizar.es, usask.ca, hil.no, unze.ba, fu-berlin.de, ecla.de, bvb.de, lsmu.com. Answers at the foot of the page.
npm install swot-bloom
import checkEmail from 'swot-bloom';
checkEmail('mv037@hdm-stuttgart.de'); // trueIf you want to generate a new bloom filter with different size/false positive rates that can be done by running:
npm run generateFilter *m/n* *k*
For further information on the parameters see Bloom Filters - the math.
- You can search by email and domain names only. You cannot search by IP.
- You don't know if the email address belongs to a student, faculty, staff member, alumni, or a contractor.
- There may be a few false positives, missing institutions... maybe even a couple of typos. Contributions welcome! Please contribute domain information to the original swot project that this is forked from.
- Using a bloom filter means that there is some chance of false positives i.e. that an email address is an institutional address when it is not. I've chosen a false positive rate of ~0.5% so the bloom filter isn't too large and can be used client-side. If this isn't acceptable, consider using swot-js. However swot-bloom is tested against 127 common non-academic email domains, non of which result in a false positive.
Please note: just because someone has verified that they own lreilly@stanford.edu does not mean that they're a student. They could be faculty, staff, alumnni, or maybe even an external contractor. If you're suddenly getting a lot of traffic from websites like FatWallet or SlickDeals, you might want to find out why. If you're suddenly getting a lot of requests from a particular school, you should look into that too. It may be good business, word of mouth, or someone may have found a loophole. Swot gives you a high confidence level - not a guarantee. I recommend putting some controls in place or at least monitor how it's doing from time to time.
Hopefully, you'll be surprised by some of this:
| Domain | Academic? | Comments |
|---|---|---|
stanford.edu |
✔️ | OK, this was an easy one so you could get at least one right |
america.edu |
✖️ | Prior to October 29th 2001, anyone could register a .edu domain name (details) |
duep.edu |
✔️ | Alfred Nobel University is a Ukranian University in the Ukraine i.e. not in the USA 🇺🇸 |
gla.ac.uk |
✔️ | Glasgow University in Scotland |
unizar.es |
✔️ | The University of Zaragoza in Spain |
usask.ca |
✔️ | The University of Saskatchewan in Canada |
hil.no |
✔️ | Lillehammer University College in Norway |
unze.ba |
✔️ | University of Zenica in Bosnia and Herzegovina |
fu-berlin.de |
✔️ | Free University of Berlin in Germany |
ecla.de |
✔️ | ECLA of Bard is a state recognized liberal arts university in Berlin, Germany |
bvb.de |
✖️ | It's a soccer team from Germany |
lsmu.com |
✔️ | Lugansk State Medical University in the Ukraine |
If you verified this by visiting all of the websites, how long did it take you? Did you have fun? Imagine you had to do this 10 - 100 times every day. Now you know a little something about the inspiration for Swot. Swot can verify them all in a fraction of a second and remove a 💩 part of someone's job.