Facebook Publishes Its AI Research To Help People Make Better Bots

Meetic Facebook Bot

Facebook has decided to open source its AI research to help people build better bots.

The social network is making the work of its Facebook AI Research lab available to everyone, posting the library of code on Github.

The fastText library is a designed to help build scalable solutions for text representation and classification.

Facebook says its research is faster than other methods, and can train more than one billion words in less than ten minutes using a standard multicore processor.

In addition to this, Facebook says it can classify half a million sentences among 312,000 classes in less than a minute.

Speaking about the research, Facebook said: “FastText combines some of the most successful concepts introduced by the natural language processing and machine learning communities in the last few decades.

“These include representing sentences with bag of words and bag of n-grams, as well as using subword information, and sharing information across classes through a hidden representation.

“We also employ a hierachical softmax that takes advantage of the unbalanced distribution of the classes to speed up computation. These different concepts are being used for two different tasks: efficient text classification and learning word vector representations.”

And the research is now publicly available on Github here.

Facebook’s AI Research team said it ultimately hopes posting the fastText research “will help us all design better applications and further advance the research in language understanding.”

Earlier this year, European dating giant Meetic built a personal dating assistant Messenger bot for singles that chats with users and helps them improve their chances of finding love.

Read more about it here.