Taxonomy Reconciliation Review¶
Structured review — instantiates Schema Conflict Resolution
Walks two hierarchical or faceted category trees node by node and decides, for each, whether it should be merged, split, translated, or preserved as a distinct branch.
When the schemas in conflict are trees — hierarchical or faceted classification systems — the resolution has a distinctive shape: it happens branch by branch, and the hard question at each node is where a category should sit. Taxonomy Reconciliation Review is the structured walk that answers it. Working node by node across two category trees, it decides for each whether to merge duplicated branches, split an overloaded one, translate between them, or preserve a branch that must stay distinct. Its defining property is that it reasons about hierarchy and placement — parent/child, breadth vs. depth, where a node belongs — rather than about entity-relations, data fields, or vocabulary. The output is a reconciled tree plus the per-node decisions and the branches ruled off-limits to collapse.
Example¶
A media company acquires a rival streaming service, and both carry a content taxonomy. The genre facets clash: the acquirer files "Documentary" as a top-level genre with Nature, True Crime, and Biography beneath it; the acquired service treats "True Crime" as its own top-level genre, cutting across Documentary and Drama. The reconciliation review walks the trees node by node. Nature under Documentary is a duplicate across both — merge. The acquirer's flat "Documentary" bucket is overloaded and hides the True-Crime demand the other service captured — split it and promote True Crime to a cross-cutting facet. A regional facet the acquired service used for local-language originals has no equivalent in the acquirer's tree and drives licensing rights — preserve it as a distinct branch with a boundary rule that it must not be folded into a generic "International" bucket. Each node ends with a recorded decision, and the collapse-forbidden branches are marked so a later cleanup can't quietly erase them.
How it works¶
- Walk node by node. Traverse both trees in parallel, comparing categories at each level rather than judging the taxonomies as wholes.
- Classify each correspondence, then decide its fate. Record how the nodes correspond, then choose the action: merge, split, translate, or preserve.
- Watch placement, not just labels. The interesting conflicts are structural — a category that is top-level in one tree and a leaf in another, or a facet that cuts across another's hierarchy.
- Mark the untouchables. Branches that encode rights, jurisdiction, or local demand get an explicit boundary rule forbidding their collapse.
Tuning parameters¶
- Merge aggressiveness — how readily near-duplicate branches are collapsed. Aggressive merging yields a tidier tree but risks flattening distinctions that earned their own branch.
- Depth vs. breadth bias — whether reconciliation favors deep hierarchies or broad faceted flats. The choice shapes navigation and search and is hard to reverse later.
- Facet independence — whether cross-cutting concepts become their own facets or stay nested under one parent. Independent facets capture multi-dimensional content but complicate the model.
- Preservation threshold — how strong a reason a branch needs to survive as distinct. A low threshold protects local meaning; a high one drives toward a leaner tree.
When it helps, and when it misleads¶
Its strength is handling the specifically hierarchical conflict — overloaded parents, cross-cutting concepts, categories at incompatible depths — that a flat correspondence table can't express. Recognizing when a concept should become an independent facet rather than a nested child is the move that most often untangles a merged taxonomy.[n1]
Its failure mode is over-integration toward one tidy tree: hierarchies invite a drive for a single clean structure, and the review can merge or delete branches that a smaller or acquired system depended on. The classic misuse is letting the dominant party's taxonomy become the skeleton and hanging the other's categories off it as afterthoughts — eliminating exactly the distinctions the smaller tree existed to capture. The guarding discipline is to make "preserve as distinct" a first-class outcome, protected by an explicit boundary rule, and to require a real downstream reason before any branch is collapsed away.
How it implements the components¶
integration_decision— each node ends in a committed action: merge, split, translate, or preserve.category_correspondence— the review records how each pair of nodes corresponds before deciding its fate.boundary_maintenance_rule— branches that must not be collapsed (rights, jurisdiction, local demand) get an explicit preservation rule.
It decides the fate of tree nodes but does not elicit the underlying entity-relation models (schema_a, schema_b) — that is Ontology Mapping Workshop — nor test the reconciled tree against real cases (validation_case_set) — that is Case-Based Mapping Test.
Related¶
- Instantiates: Schema Conflict Resolution — the hierarchy-specific resolution that reconciles two category trees node by node.
- Sibling mechanisms: Boundary-Spanner Review Session · Case-Based Mapping Test · Data Mapping Specification · Glossary Alignment Table · Ontology Mapping Workshop · Schema Crosswalk Table · Semantic Interoperability Review · Translation Register
Editorial Notes¶
Form Classification¶
Form family: Assessment, Review & Assurance
Rationale: Taxonomy Reconciliation Review operates as a bounded evaluation of existing evidence or work that produces a finding or disposition because it walks two hierarchical or faceted category trees node by node and decides, for each, whether it should be merged, split, translated, or preserved as a distinct branch.
Independent corroboration: The frozen evidence defines Taxonomy Reconciliation Review as 'Walks two hierarchical or faceted category trees node by node and decides, for each, whether it should be merged, split, translated, or preserved as a distinct branch', so its operative form is Assessment, Review & Assurance.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Library & Information Science
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Universal
Rationale: Taxonomy reconciliation review derives most directly from library and information science's classification, provenance, and retrieval tradition; its defining operation is to walks two hierarchical or faceted category trees node by node and decides, for each, whether it should be merged, split, translated, or preserved as a distinct branch.
Related originating lineages:
- Computer Science & Software Engineering — Computer science and software-engineering practice supplies a parallel or contributing lineage for the mechanism's defining operation: walks two hierarchical or faceted category trees node by node and decides, for each, whether it should be merged, split, translated, or preserved as a distinct branch.
- Linguistics & Semiotics — Linguistics and semiotics' terminology, meaning, and sign-system tradition provides a formative adjacent lineage for the same taxonomy reconciliation review operation.
Review resolution: Both blind reviewers independently select library_information_science as the primary historical origin for the concrete operation—Walks two hierarchical or faceted category trees node by node and decides, for each, whether it should be merged, split, translated, or preserved as a distinct branch. The queued differences concern alternate origin disagreement, origin mode disagreement, domain reach disagreement, encyclopedia synthesis disagreement, not the primary lineage. I retain every alternate that either reviewer explains, without a numeric cap, and choose origin_mode=cross_disciplinary_synthesis because the reviewers' combined evidence identifies material construction from multiple disciplines. domain_reach=universal records later portability rather than multiplying historical origins; confidence=high is the conservative shared evidentiary level, and encyclopedia_synthesis=true preserves either reviewer's affirmative synthesis finding.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
[n1] Faceted classification, associated with S. R. Ranganathan, describes content by several independent attributes (facets) rather than forcing it into one strict hierarchy — the reason a cross-cutting concept like "True Crime" is often better modeled as a facet than as a child of a single parent. ↩