Whilst Wikipedia exists in hundreds of different languages, the coverage in each of those languages is pretty variable to say the least.  Whilst there is reasonable coverage of content between individual languages, things get much murkier when you get into local differences.

A recent study highlighted how editors can be reminded of the important gaps that are missing in the coverage provided by the site.

Plugging the gaps

The team of researchers created an easy to use recommendation tool that highlights the key articles that thus far do not have coverage in a particular language.  If they have a skill in this particular language, they can then look to plug that gap with a new piece.

So, for instance, it would be capable of first identifying a gap in the database, and then find an editor that has a history of editing in that subject area, and who also has the relevant language skills, before reaching out to them to recommend they do an article on that topic.

“As university researchers, we look for projects with real-world impact,” the authors say. “What could have more impact than democratizing access to knowledge?”

Looking for links

The project was possible due to the huge amount of information contained by Wikipedia about both the different languages articles are written in, and the connections between articles.

They began by listing each article in every language, before then cross-referencing these lists to try and find gaps in certain languages.  They next attempted to determine the importance of that article in that region to create a ranking for the most important missing articles.

“We had to create a system of rankings that would be meaningful to editors in different cultural and linguistic communities because Wikipedia is shaped by the editors’ choices,” they say.

The aim is to give editors the greatest bang for their buck in terms of the effort they put into the site.  They tested their system out on nearly 5 million articles from the English version of Wikipedia.  When this list was analyzed, it emerged that there were roughly 1.6 million of those articles missing a French alternative.

After identifying the 300,000 most important missing pieces, the group was split into three and sent to a select group of 12,000 editors that had previously edited in both English and French Wikipedias in the last year.

Each editor was directed to five of the missing articles with a suggestion that they could make a real difference by contributing a missing piece.

A helpful nudge

A month after the emails were sent out, the researchers re-spidered the database to see how many of the gaps had been plugged.  They found that with a carefully targeted email, they could triple the normal rate of article creation.

Such has been the success of the tool, the Wikimedia Foundation have developed a tool that aims to perform a similar function and help editors find gaps in their local language Wikipedia, and then suggest they translate that into another language.

Hopefully, it will go a long way towards filling in some of the linguistic gaps that are still present in the site.