Semantics
Semantics is the meaning and behavior conveyed or assigned to expressions, constructs, representations, or their combinations in a language or symbolic system.
LLMs in Production
Semantics is the study of meaning in language, including the meanings of expressions and their combinations.
- How it is used
- Distinguishes meaning from fluent syntax when assessing what LLM outputs actually convey.
- How deep it goes
- Explains semantic approximation through embeddings and a vector analogy, while stopping short of claiming embeddings create human meaning.
- Examples
- Uses a Word2Vec king–man+woman analogy, an “I need food” context example, and an embedding plot.
- What you can do after
- Evaluate/understand whether fluent output or nearby embeddings support a meaning claim without treating either as proof of understanding.
Knowledge Graphs and LLMs in Action
Semantics is the explicit representation of the meaning of data through typed entities and meaningful relationships.
- How it is used
- Makes knowledge-graph meaning explicit so data can support contextual analysis, business decisions, explainability, and reasoning.
- How deep it goes
- Explains typed relationships, taxonomies, ontologies, and pragmatic partial modeling without detailing formal implementation procedures.
- Examples
- Uses Vehicle–Car–Sedan hierarchies, Car–Automobile identity, Car–Bicycle disjointness, and healthcare specialty ontologies.
- What you can do after
- Evaluate graph representations for meaningful relationships, explainability, and extensible domain modeling.
Read how Knowledge Graphs and LLMs in Action explains semantics →
Related concepts
- natural languagepart of · Natural language is a human language that develops through social use and is used for communication, in spoken or written form.
- syntaxcontrasts with · Syntax is the set of rules governing how elements of a language or notation may be arranged to form valid expressions or constructs.
- embedding layeruses · An embedding layer maps discrete identifiers, such as tokens or categories, to learned dense numerical vectors.
- Pythonused by · Python is a high-level, general-purpose, dynamically typed programming language known for readable syntax and automatic memory management.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.