Disambiguation
Disambiguation is the process of resolving ambiguity by selecting the intended interpretation, entity, or goal among plausible alternatives.
Knowledge Graphs and LLMs in Action
Named entity disambiguation is the technique of determining which real-world entity a mention in text refers to and linking it to the correct knowledge-base entity.
- How it is used
- Uses contextual and ontological evidence to ground ambiguous mentions for biomedical knowledge graphs and decision-support analysis.
- How deep it goes
- Explains candidate selection, ranking, ontology integration, ingestion, and LLM path-based disambiguation, while noting traditional-system limitations.
- Examples
- Uses Zika mentions, UMLS identifiers, scispaCy rankings, SNOMED relationships, Python ingestion, and Neo4j knowledge-graph construction.
- What you can do after
- Build it with scispaCy and UMLS, ingest linked entities into Neo4j, or operate an ontology-guided LLM pipeline for complex biomedical text.
Read how Knowledge Graphs and LLMs in Action explains disambiguation →
Effective Conversational AI
Disambiguation is a conversational technique that presents alternative interpreted goals so a user can select or clarify the intended one.
- How it is used
- Clarifies a user’s intended goal or response when a chatbot has several plausible interpretations.
- How deep it goes
- Defines classifier-based intent choices and illustrates a conversational follow-up, without specifying an implementation.
- Examples
- Contrasts a retry with a reassuring text-or-phone follow-up and uses a booster-and-appointment request to show multiple intents.
- What you can do after
- Practice/perform disambiguation by offering relevant choices instead of retrying a good-faith response.
Read how Effective Conversational AI explains disambiguation →
AI-Powered Search
Query disambiguation is the process of resolving multiple plausible interpretations of a search query.
- How it is used
- Resolves ambiguous search terms into contextual senses before choosing how to present results.
- How deep it goes
- Works through graph traversal, configurable limits, and relatedness scores, but offers result-presentation strategies as options rather than an implemented ranking policy.
- Examples
- Compares travel and DevOps senses of “driver” and cooking and DevOps senses of “chef” using related terms and category scores.
- What you can do after
- Apply it via a semantic knowledge graph to identify competing senses and choose grouping, diversity, or alternative query suggestions.
Related concepts
- candidate disambiguationhas part · Candidate disambiguation is the phase of named entity disambiguation that uses contextual and ontology-based evidence to select the intended entity from candidate entities.
- named entity recognitioncontrasts with · Named entity recognition is a natural-language-processing task that identifies textual mentions of entities and classifies them into categories such as people, places, organizations, or dates.
- candidate rankinghas part · Candidate ranking is the phase of entity disambiguation that scores and orders possible reference entities using contextual and other evidence.
- candidate selectionhas part · Candidate selection is the phase of entity disambiguation that identifies possible reference entities for a recognized mention.
- knowledge baseuses · A knowledge base is an organized collection of information that can be accessed or queried to support reference, retrieval, or reasoning.
- contextuses · Context is the surrounding information, state, or circumstances that condition how an entity, event, or task is interpreted or how a system behaves.
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.