Search engines: bridging the gap for minority languages

We have all looked to “Dr. Google” for a health problem or to find a recipe or to learn how to fix something maybe. Sometimes helpful, sometimes not. But what if you ask Google for something? and that it doesn’t even recognize your language?

Phil Parker

Phil Parker, talk to COSM 2021, told the story of a woman in Ethiopia looking for a ‘breast lump’, using one of the more than 80 languages ​​or dialects spoken in the area. Its language was one of thousands spoken by a relatively small population.

The search engine did not recognize his entry and returned no results. She tried her query in Swahili, but found nothing instructive about “breast lumps” in Swahili. She eventually tried her search in English. Several hits appeared. Indeed, she had to understand English to access content on the web.

More than a third of the perhaps 6,000 languages ​​in the world are spoken by less than 100,000 people. (Unesco) Twenty-five percent of the approximately 6,000 languages ​​in the world are spoken by less than 1,000 people. (BBC) In total, that’s a lot of people who can be excluded from vital information online.

Is there a way for this woman to have access to information in her own language?

Parker, INSEAD Full professor of Management Sciences and founder of Botipede, presented his solution to COSM 2021 conference last week. His system combines a decentralized “content engine” with natural language understanding.

In previous work, Parker has developed an automated system that translates textbooks and other educational materials into languages ​​spoken by small populations. This work caught the attention of the Bill and Melinda Gates Foundation, which funded a project using Parker’s automated authoring processes and translating agricultural and local weather information. Today, around ninety countries have access to its weather platform.

In his presentation, Parker presented his idea for algorithmically filling the translation gap for all other content on the web. Botipedia uses natural language learning and algorithmic search engine filtering to create what it calls a “content engine.” He predicted that this kind of decentralized and privatized system will be the future of the Internet after Google: the file will exist.

Here is a more detailed explanation of Botipedia:

You can also read: If Google thinks for you, use THEIR search engine. Otherwise… Google’s monopoly affects the free exchange of ideas in the public arena and our electoral process. Brave Search offers the first real alternative to Google since Bing by introducing a third index in English and protecting user privacy. (Caitlin Bassett)


Source link

Rosemary S. Bishop