A surprising amount of work goes into coming up with the nearly-nonsensical names for drugs, and the process often starts with a list of thousands(!) of candidate names which are winnowed down. For example, generic name needs to follow certain rules: the suffix indicates the class of drug (see the giant list here[0]), and there are other constraints about what must (and must not) go in the prefix. Brand names have some regulatory restrictions, but are also optimized for marketing: Viagra, for example, was supposedly named by combining 'vigorous' and 'Niagara [Falls]', which is...connotative of its effects. Something like this could be helpful for noodling around in that space.
It could be useful for naming all kinds of things.
On the other hand, the rest of the stuff on that page is all fairly random and not particularly tied to Google's core products. It might just be a random side project.
"Serious play means you value play as a tool for fostering creative thinking."
This could be the seed, test-bed, and proof-of-concept for a set of libraries that would be useful for speech to text, autocomplete, natural language processing, etc...
14 comments
[ 2.2 ms ] story [ 45.9 ms ] threadhttps://www.youtube.com/watch?v=-VsmF9m_Nt8
[0]: https://druginfo.nlm.nih.gov/drugportal/jsp/drugportal/DrugN...
On the other hand, the rest of the stuff on that page is all fairly random and not particularly tied to Google's core products. It might just be a random side project.
"Serious play means you value play as a tool for fostering creative thinking."
This could be the seed, test-bed, and proof-of-concept for a set of libraries that would be useful for speech to text, autocomplete, natural language processing, etc...
It uses Markov chains to pick the next letter (the current state is the previous two letters)