NLP & Linguistics Overview

[Literal] Survey post on natural language processing and linguistics: language as symbolic signaling, plus core subfields and modern vector methods. [AI Synthesis] Maps classical layers (sound → meaning → context) onto embedding and LM practice.

Key points

• [Literal] Human language is symbolic signaling; words map signifiers to signified ideas or things.
• [Literal] Phonology studies sound patterns—stress via pitch, loudness, or length; intonation reflects syntax/semantics.
• [Literal] Morphology covers word formation; syntax, semantics (incl. lexical), and pragmatics (context) complete the stack.
• [Literal] Word vectors encode context-based meaning; distributional similarity motivates Word2Vec-style neighbors.
• [Literal] Language modeling is a probabilistic model of word sequences.

Patterns / reminders

• [AI Synthesis] When stuck on NLP tasks, locate which linguistic layer (sound, form, meaning, context) the error belongs to.

Sources

• (Source: raw/_posts/2017-05-19-nlp.md)