Created by August Edström
about 6 years ago
|
||
Question | Answer |
Why is natural language important in AI? | -Store,transfer,interact with language -Loads of available data -AI can make interesting use of this data |
What is lingustic rationalism? | Knowledge is not just gained through senses but humans also have "an innate system of language that facilitiates a rational mechanism of language when receiving lingustic input"(Chomsky) |
What is linguistic empiricism? | Lingustic input + cognitive abilities. Learn by a lot of input basically |
What kind of model is Categorical Grammar? | A non-probabalistic model |
What techniques can be used in probabalistic models? | N-gram model (a Markov chain) |
What is competence?(according to Chomsky) | Our internal knowledge of language |
What is performance?(according to Chomsky) | The actual lingustic behaviuor |
What is a language model? | A set of strings -set of sounds -set of words -set of sentences Modelling language - computing set membership |
What is a N-gram? | A sequence of n words |
What does P(w|h) say? | The probability of a word w given some history h |
What is lexicalised grammar? And is it a non-probablistic model or probabilistic model? | Lexicalised grammars do away with the rule-based nature of grammar Idea: see some leaves (words) are functions that take other leaves to form nodes. A non-probabilistic model |
What is sentiment analysis? | Analyze a piece of text and rate it according to different categories |
Want to create your own Flashcards for free with GoConqr? Learn more.