Entity Resolution
Definition
Entity Resolution is the process of determining whether different records, references or representations refer to the same real-world entity and consolidating or linking them appropriately.
Why it matters
Entity resolution is important when an organisation, product or person appears across multiple sources with different names, identifiers, attributes or levels of detail. Weak resolution can fragment entity signals or merge unrelated entities.
How it works
Record comparison
- The system compares names, identifiers, attributes, relationships and contextual signals across records or mentions.
Match decision
- Records are classified as matching, non-matching or unresolved based on available evidence.
- Exact matching may be sufficient for strong identifiers, while ambiguous cases require broader context.
Canonicalisation
- Confirmed matches can be linked to a canonical entity or master record.
- Conflicting attributes may still need separate resolution or source authority rules.
How Netsleek uses the term
Netsleek uses Entity Resolution as the broader identity process that explains how brand, product, parent-company and third-party references can be reconciled across the web and knowledge systems. It is distinct from simply detecting or linking one text mention.
Comparisons
- Entity Resolution vs Entity Linking: Entity Linking maps a mention to a canonical entity. Entity Resolution reconciles multiple records or representations that may refer to the same entity.
- Entity Resolution vs Entity Disambiguation: Disambiguation chooses between competing candidate identities. Resolution can involve broader record matching and consolidation.
- Entity Resolution vs Canonical Entity: Resolution is the process. The Canonical Entity is the resulting reference identity.
How Entity Resolution Connects Across AI Systems
Entity Resolution does not operate in isolation. It connects to other established concepts across the AI search and retrieval process:
- Uses entity signals: Names, identifiers, locations, relationships and other Entity Signals can support matching.
- Supports canonicalisation: Resolved records can be associated with a Canonical Entity.
- Supports knowledge graphs: Consistent entity identities improve graph relationships and knowledge integrity.
- Affects retrieval: Fragmented or confused identities can reduce entity-aware retrieval and recommendation quality.
Related glossary concepts
- Entity Linking
- Entity Disambiguation
- Canonical Entity
- Entity Mapping
- Entity Signals
- Entity Confusion
- Knowledge Graph
Common misinterpretations
- Entity resolution is not simply duplicate removal.
- Two similar names do not automatically represent the same entity.
- Resolution can remain uncertain when identifiers and contextual evidence conflict.
Summary
Entity Resolution determines when different records or representations correspond to the same real-world entity. It supports canonical identity, knowledge integrity and consistent entity-aware retrieval across heterogeneous sources.