Introduction to Natural Language Processing (자연언어처리)

Introduction to Natural Language Processing (자연언어처리)

1. definition

   • any language that has envolved naturally in humans through use and repetition without conscious planning or premediatation

2. Goal

   • to specify models that approach human performance in the linguistic tasks of reading, writing, hearing, and speaking

3. Steps of NLP

   • Morphology
     • Concerns the construction words
   • Syntax
     • Grammar
       • a formal spesification of the structures allowable in a language
     • Parsing Technique
       • a method of analyzing a sentence to determine its structure according to the grammar
   • Semantics
     • from syntactic analysis to a logical representation
   • Speech Act
     • Assert
       • the proposition is being asserted
     • Y/N-Query
       • the proposition is being queried
     • Command
       • the preposition describes an action to perform
     • WH-Query
       • the preposition describes an object to be identified
   • Discourse
     • concerns how preceding sentences affect the interpretation of the next sentence
   • Pragmatics
     • concerns how sentences are used in different situations and how the use affects the interpretation of the sentence

4. main problem of NLP

   • every NLP step has its own ambiguities
     • Morphological Analysis
     • POS tagging
     • Syntactic Ambiguity
     • Semantic Ambiguity

5. resolve ambiguities

   • Corpora (Computer-readable Collection of Text or Speech)
     • generate knowledge automatically with statistical methods and Machine learning
     • image captioning
     • word representations
       • one-hot(local) representation
     • Word-Net
       • nouns, verbs, adverbs and adjectives are grouped into sets of cognitive synonyms
     • Distributed Representation
       • represents words as real-valued dense vectors
     • Word Embedding
       • symbol(word) -> continuous vector
     • Distributional Hypothesis
     • Word2Vec

Imágenes y archivos del mapa