Candidate disambiguation

As explained in Knowledge Graphs and LLMs in Action

Candidate disambiguation is the phase of named entity disambiguation that uses contextual and ontology-based evidence to select the intended entity from candidate entities.

Candidate disambiguation is the final named entity disambiguation phase: it selects the intended entity from generated candidates by combining co-occurring context with domain-ontology evidence. section 8.4 section 8.4.3

A practical candidate-disambiguation pipeline proceeds as follows:

• Detect shortest paths between candidates associated with different mentions in the sentence. section 8.4.3

Zika and microcephaly

When a sentence mentions “Zika” with “microcephaly,” their co-occurrence favors Congenital Zika virus infection over a general virus interpretation; the resulting selections can be returned with SNOMED identifiers. section 8.4.3

Figure 8.6 NED CD steps: (1) detecting the shortest paths between all the candidates related to the entity mentions in the sentence; (2) translating detected paths into natural language sentences; (3) summarizing those sentences into useful text for disambiguation

The workflow diagram shows the three operational steps: shortest-path detection, path translation, and textual summarization. section 8.4.3

How Knowledge Graphs and LLMs in Action covers it

How deep it goes

Explains a three-step graph-to-text pipeline and final contextual selection, using shortest paths, translation, and summarization.

Examples

Uses Zika, microcephaly, congenital malformations, SNOMED paths, Neo4j queries, translated sentences, and structured disambiguation results.

What it assumes

Assumes identified mentions, generated candidate entities, a structured domain ontology, and an LLM capable of processing textual context.

What you can do after

Apply it via Neo4j Graph Data Science and LLM prompts to produce ontology-linked entity choices with contextual support.

How it connects

Connects named entity disambiguation with candidate selection, knowledge graphs, domain ontologies, large language models, and shortest-path algorithms.

liveBookmanning.com

Read Knowledge Graphs and LLMs in Action on liveBook Get the book at manning.com

Related concepts

• disambiguationpart of · Disambiguation is the process of resolving ambiguity by selecting the intended interpretation, entity, or goal among plausible alternatives.
• path-to-text translationhas part · Path-to-text translation is the conversion of a graph path and its relationships into a natural-language textual representation.
• shortest-path detectionhas part · Shortest-path detection is the algorithmic process of finding a path between nodes with minimum total length or cost.
• textual path summarizationhas part · Textual path summarization is the condensation of a graph path’s relationships and entities into a concise natural-language description.
• shortest pathuses · A shortest path is a path between two nodes or locations that minimizes total distance, length, or cost under a specified measure.

This explainer was generated by Manning's knowledge graph from the text of Knowledge Graphs and LLMs in Action. The section links point to the parts of the book it draws on.