DataTerminology.comNEW
Also known as: Sense
A distinct meaning of a word in context, often one of several polysemous interpretations.
DESCRIPTION
Word senses capture the specific meanings words take in particular contexts, central to lexicography and NLP. Many words are polysemous, requiring models and rules to determine the correct sense. Sense inventories like WordNet list senses per lemma and link them to synsets representing concepts. Word sense disambiguation algorithms infer senses using surrounding context and knowledge resources. Accurate sense handling improves machine translation, search, summarization, and information extraction.
ALTERNATIVE DEFINITIONS
SIL Glossary of Linguistic Terms
Sense is one of the meanings of a lexeme.
Defines sense succinctly within linguistic terminology.
Wikipedia
In linguistics, a word sense is one of the meanings of a word.
General-purpose definition highlighting polysemy and lexicographic context.
NLTK Documentation
Each word can have several senses; each sense is a member of a synset.
Emphasizes relation of senses to synsets in WordNet.
RELATED TERMS
WordNet Polysemy Homonym Word Sense Disambiguation Lemma Lexeme BabelNet
COMPETING TERMS
KNOWLEDGE AREAS
AI Data Science Semantic & Knowledge Modeling Metadata Management
TARGET AUDIENCE
AI Engineer Data Scientist Data Modeler Data Steward
EXTERNAL SOURCES
- Wikipedia - Word sense
- Princeton WordNet - Sense inventories and synsets
- NLTK - Programmatic sense access
- BabelNet - Multilingual sense-concept mapping
TOP QUESTIONS
How does Word Sense relate to Synsets?
Each word sense is a member of a synset that represents a concept. Synsets aggregate senses across different words sharing the same meaning.
Category: Misunderstanding
Why is Word Sense Disambiguation important?
It enables accurate interpretation for tasks like translation, search, and extraction. Misclassification of senses leads to semantic errors and degraded performance.
Category: Relevancy
What causes multiple Word Senses?
Polysemy, metaphor, domain-specific usage, and idiomatic expressions lead to multiple senses. Context determines which sense is intended.
Category: Overview
What are common mistakes with Word Sense modeling?
Assuming dictionary sense granularity matches application needs. Overly fine-grained senses reduce reliability; consider sense clustering.
Category: Mistakes
Which resources catalog Word Senses?
WordNet, BabelNet, and domain dictionaries catalog senses and link them to concepts. Specialized glossaries capture domain-specific senses.
Category: Resources
How do modern LLMs handle Word Senses?
LLMs infer senses from context embeddings. Integrating lexical resources can improve explainability and robustness.
Category: Topic
How to evaluate Word Sense Disambiguation?
Use benchmark corpora and metrics like precision, recall, and F1. Human annotation and sense inventories support evaluation.
Category: Testing
Where to start learning about Word Senses?
Study WordNet documentation and NLTK tutorials, then explore WSD literature and BabelNet for multilingual perspectives.
Category: Learning
What is the risk of ignoring Word Senses in NLP?
Ambiguity can propagate errors into downstream tasks, harming user experience and decision accuracy.
Category: Risk