HAC-E network clustering
Algorithms AIMech:HACE DRAFT
Aliases: HAC-E
A hierarchical network-clustering variant that greedily guides agglomeration by the observed edge density between two vertex groups and uses the resulting group hierarchy for link-probability estimation.
Identity Notes & Concept Scope
Compared all 442 pinned AIO and 669 native IDs, names, aliases and scoped synonyms for the three candidates and publication identifiers; no normalized identity match. Ignored- and hidden-inclusive fixed-string searches of the recovery, older discovery and original Documents checkouts returned 158, 0 and 5 lines. All 158 public matches were inspected and belong to source/discovery/review context and rendered existing records, not these maintained identities. The five original-checkout matches are unrelated Spanish hace text in ignored dependencies. Semantic assessment covered 46 AIO and 264 native definitions: 44 AIO and 260 native rows were unchanged from prior reading; two additional AIO and four additional native definitions were read. Sixteen complete annotated AIO node/edge/generated contexts and the candidate-specific native records were assessed. Retain the source-named HAC-E network variant separately from HAC-ML, HAC-ES and HAC-Q. Its merge score is mathematically average binary adjacency, not a novel linkage formula. The named graph-model and link-prediction use is narrower than generic average linkage; no current maintained identity covers it.
Claims & Grounded Evidence
4 Claims
Mechanism
#edge-density-guide
The merge guide is the observed number of edges between two disjoint nonempty vertex groups divided by their number of possible cross-group pairs. Replacing HAC-ML’s profile-likelihood guide with this density gives HAC-E. The score favors dense cross-connections and does not directly compare agreement with all external groups.
Scope
#average-linkage-boundary
For a binary adjacency matrix, the density is exactly mean cross-group adjacency; its complement is average dissimilarity when pair dissimilarity is one minus adjacency. The score after a merge therefore uses group-size weights, rather than an unweighted mean of the two old scores. This algebra identifies a shared linkage criterion, not equivalence of every implementation, search restriction or stopping rule.
Scope
#score-versus-prediction
The density-based merge guide is distinct from the hierarchical model’s later link estimates. The study estimates densities between left and right vertex groups at tree joins for hierarchical methods. A score comparison alone does not establish a calibrated probability of a missing edge, a complete merge sequence or recovery of true biological communities.
Scope
#evaluation-scope
The study compares this named variant with HAC-ML and other methods on biological interaction networks using held-out-edge prediction. Positive test pairs are balanced by an equal number of sampled holes, and maximum F scores summarize ranked predictions. This selection changes the negative prevalence relative to deployment. The reported comparisons support a setting-specific empirical assessment, not universal superiority or calibrated real-world link probabilities.
⚠️ Curated Limitations & Negative Results
- The average-adjacency criterion is mathematically related to average linkage. This record preserves the source-named network variant without claiming a novel general linkage function.
- On a three-vertex path, density favors adjacent singleton pairs while the HAC-ML profile guide favors the nonadjacent pair with identical external connectivity. This is a local score distinction; candidate restrictions and a complete execution were not modeled.
- This record covers a named 2011 score variant in Park and Bader’s network-clustering study, not a fitted interactome or the entire family of agglomerative clustering methods.
- The paper defines each alternative by replacing the merge guide. It does not separately specify every candidate-restriction, tie, stopping and terminal-collapse convention for each variant. The detailed HAC-ML implementation and its complexity are not automatically asserted for this variant.
- The experiments use undirected, unweighted interaction graphs with iterative removal of isolated and degree-one vertices. Missing observed edges can include unobserved biological interactions; sampled holes are not established biological negatives.
- No author implementation, data or supplementary archive was audited or executed. Finite constructed checks assess local scores and algebraic distinctions, not complete clustering runs, benchmark replication or global optimality. DRAFT remains.
Artificial Intelligence Ontology (AIO) Grounding
Mapped AIO Classes
| AIO Target Class | Mapping Relation | Mapping Rationale |
|---|---|---|
AIO:HierarchicalClustering
Hierarchical Clustering
|
related_match | This algorithm constructs a hierarchy of vertex groups. The target is a related task, not an equivalent algorithm or asserted superclass. |
AIO:UnsupervisedClustering
Unsupervised Clustering
|
related_match | The procedure groups graph vertices using observed relationships rather than supervised label prediction. The task relation does not establish algorithm identity. |
Evidence Sources & Primary Literature
1 Source| Source ID | Publication / Document | Type / Access | Reference & Checksum |
|---|---|---|---|
park2011
|
Yongjin Park, Joel S Bader
|
primary paper
full_text
|
ID: doi:10.1186/1471-2105-12-S1-S44
sha256:e930abd762f0…
|
Curation Provenance
Curator: Codex
Method: ai_assisted
Model: GPT-6
Created: 2026-10-14T01:20:31Z
Updated: 2026-10-14T01:20:31Z
Source File:
data/curated/HACE.yaml