Machine learning tool Quicksilver can find overlooked females

Wikipedia logo.

Primer's John Bohannon has been discovering people's work and contributions thanks to a machine learning system built at Primer. "It does this much as a human would, if a human could read 500 million news articles, 39 million scientific papers, all of Wikipedia, and then write 70,000 biographical summaries of scientists." The reason this is news is because you would probably never know about their contributions by frequenting Wikipedia, but he knows a way to fix that.

Wikipedia appears have a gender problem, for one. It is a matter of under-representation. And now the machine learning system at an AI startup has been showing how it could address the situation.

Primer is in the news. The Primer system was trained on scholarly journals. The gender gap-filling tool is called Quicksilver. It can spot many an overlooked female scientist with no presence at Wikipedia. Cory Doctorow in Boing Boing said 18% of Wikipedia's biographic entries were about women and the vast majority of Wikipedians were men.

The show-all process involved 30,000 Wikipedia entries to create a model that allowed it to identify characteristics making a scientist noteworthy for encyclopedic inclusion. Then, it mined the academic search-engine Semantic Scholar to identify 200,000 authors of .

Tom Simonite said in Wired: "Only 18 percent of its biographies are of women. Surveys estimate that between 84 and 90 percent of Wikipedia editors are male."

Actually, if you catch the story in Wired, the gender fix is part of the bigger story of Quicksilver looking for unsightly gaps.

In the bigger picture, blogged Bohannon, "Our aim is to help the open data research community build better tools for maintaining Wikipedia and Wikidata, starting with scientific content.

(In addition, "Quicksilver doesn't just spot overlooked individuals and generate draft articles. It can also be used to maintain Wikipedia entries and identify when they haven't been updated for a while," said James Vincent in The Verge.)

So, what is the fix? Note that Primer is not about automated fixer-uppers. Said Simonite, "it doesn't plan to ever let Quicksilver autonomously add to the site." Wired quoted the CEO of Primer, Sean Gourley. "There are always humans in the loop." Popular Science said, "Quicksilver discovers scientists who should have Wikipedia articles about them and writes a first draft."

Their work continues. Bohannon said they have been quietly testing and improving Quicksilver for months. "Even before we finished the text generation component, Quicksilver was used in three English Wikipedia editathons for improving coverage of women of science. (Thank you to 500 Women Scientists for collaborating and inspiring us!)" He said they will describe their architecture in detail in future posts.

Meanwhile, wrote Simonite, "Wikipedia's notoriously punctilious community will likely keep a close eye on content generated with Quicksilver's help. One question is whether this tool aimed at fixing blind spots has any of its own."

Explore further

Most Wikipedia health articles contain errors

© 2018 Tech Xplore

Citation: Machine learning tool Quicksilver can find overlooked females (2018, August 10) retrieved 23 August 2019 from
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.

Feedback to editors

User comments

Aug 10, 2018
...and make sure they never start a family

Aug 11, 2018
why women do not edit wikipedia? Who is stopping them from doing so? Who is to blame?

men, of course....

but they are very active in facebook, I do not think there is a gender gap in the gossip sites on the internet...

Aug 11, 2018
The difference between female gossips and male gossips? Is that the males use the word 'gossip' as one of a multitude of insults against females.

Yet if a female accused a male of being a gossip? She would be lucky to escape with his fist to her mouth.

The problem is due to active misogyny, violent threats and outright hostility among male colleagues.

Many women choose to remain anonymous out of practical need to protect themselves and their families.

Now women writers use their initials or even a male pseudonym to avoid the males who will never achieve manhood.

However, that is much more difficult for women in business or the sciences. They are dependent on the infrequent good-graces of the males dominating investments, grants,product placement, promotions, award systems, professional associations and logistics.

The complaint I hear the most from women id that men are unable to keep their promises or commitments.

"It's just not manly to not cheat!"

Aug 11, 2018
So why do they are way more active in gossip sites like facebook than in wikipedia?

Oh, I got it... because if they writte on wikipedia evil men will beat them up!

But they do not get attacks for gossiping in facebook!

I am very sarcastic.... just to be sure

Aug 12, 2018
mqr, among the differences between male and female is the method of communicating.

Males are online as part of their testosterone driven need for dominance games. That is not a criticism, just an observation.

It is when those dominance games spill over into real life relationships resulting in physical violence against females, that is the point of criticism.

I urge you to talk to female authors. You will hear an earful of anger about male misogyny groups like the Sad Puppies and other altright gangs. Busily trying to sabotage the incomes of female authors.

You don't want to hear it? Makes you uncomfortable to confront ugly reality?

Well, I'm sure you can escape back into the video-game world of male dominate fantasies.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more