Women often come second to men in search engine results : journalists, writers, actors, artists, musicians, business people … However, since that 2016 article on film makers, it seems Google has tweaked its algorithm : it would be hard to search ‘film director’ without having a picture of a female film director included in the results.
Recently at the OECD Forum, Rand Hindi, the founder of AI start-up Snips said ‘Bias in Artificial Intelligence is the real risk […] but an AI remains a machine : you can control it, or just unplug it’.
How willing are we to take control of the algorithms? Should algorithms be bias-free, or should they reflect natural human biases in an un-biased way ?
Algolia ‘Search as a Service’ makes it possible to make your own search engine, tweak the ranking algorithms and apply it to anything (from e-commerce web sites to films, cultural products, … )
Search IMDB 5000 film directors with a gender angle
So we’ve played with the idea : what if you could build your own search engine with an inverted gender bias ? We’ve connected Algolia to NamSor API, made a prototype to search IMDB top 5000 movies.
Click to check-it out and search!
If you are in Paris this week-end, feel free to join Reporters d’Espoirs hackathon, as we would like to apply this idea to #PositiveJournalism. We’d like to index a large corpus from the press, use machine learning to tag articles and then allow to search with a gender angle (both journalists and cited names in articles). This has been done already by Sweden’s Rattviseformedlingen in a dashboard format. It would be nice to how tweaking the search actually impacts the way we browse and read articles.
NamSor™ Applied Onomastics is a European vendor of sociolinguistics software (NamSor sorts names). NamSor mission is to help understand international flows of money, ideas and people. We proudly support Gender Gap Grader.
Reach us at: firstname.lastname@example.org