Splitting text into words or subword pieces is called what?
AnswerTokenization
Tokenization breaks text into units a model can process.
AI · Facts & stats
NLP facts & stats
7 fact-checked facts about natural language processing and how computers understand text, each with the reason behind it. 5 more are in today's round and join this page after it closes.
The basics
Which 1950 test judges machine intelligence through conversation?
AnswerThe Turing Test
Alan Turing proposed the imitation game in 1950.
Going deeper
Who created the early chatbot ELIZA?
AnswerJoseph Weizenbaum
Joseph Weizenbaum built ELIZA at MIT in the mid-1960s.
What does stemming do?
AnswerReduces words to a root form
Stemming turns words like running and runs toward the stem run.
The BERT language model was developed by which company?
AnswerGoogle
Google researchers released BERT in 2018.
Expert level
What does BERT stand for?
AnswerBidirectional Encoder Representations from Transformers
BERT reads context from both directions using Transformer encoders.
ELIZA's best-known script imitated which kind of conversation partner?
AnswerA Rogerian psychotherapist
The DOCTOR script reflected users' statements back as questions.
Know these cold?Today's round is five questions like these, 20 seconds each. Your streak and level follow you to every arena.Play today's round