Skip to content
modumatics Modular Infrastructure for Inclusive Housing Tran Thien Toan Ngo · PhD Dissertation

Appendix: Chapter 5 Reference Tables

These tables are referenced from Chapter 5 (the standardisation schema). They were relocated from the chapter main text to keep the in-text body within the length band while preserving the full data (no content cut). All figures are census quantities over the serialised SDA Design Standard, reported descriptively with explicit denominators.

Analytical methods (Section 5.4)

The six analytical methods, each targeting a distinct dimension of ambiguity.

Method Ambiguity Dimension Instrument Output
Integrity-gate analysis Referential identity Duplicate-identifier detection across 611 records Count of divergent-text duplicates; category distribution
Multi-method polysemy protocol Lexical plurality Three-method agreement: WordNet inventory, Lesk disambiguation, embedding-context dispersion Polysemous proportion (descriptive); confidence tiers (strong / moderate / weak)
Trigram / POS structural analysis Clause-role signalling Part-of-speech trigram extraction across clause classes Class-conditioned dominant patterns; structural ambiguity exposure by class
Deontic-force classification Normative-force variation Modal-verb detection and force-type assignment (obligatory, permissive, advisory, prohibitive, mixed) Force distribution by clause class; mixed-force identification
Foundational-mapping coverage Representational completeness Term mapping against the stratified entity vocabulary (seven primitives, seven composites) Coverage ratio; fallback-to-element count; effective coverage by threshold
Figures-channel integration Cross-channel completeness Triple decomposition, polysemy detection, applicability mapping, bidirectional cross-validation Channel-specific coverage ratios; figure-only and text-only proportions

Baseline dimension summary (Section 5.11)

The six empirical-baseline dimensions in compact form.

Baseline Dimension Result Interpretive Risk
Corpus completeness 611 text records and 189 figure design requirements; no missing required keys; no truncation hits Validates that ambiguity findings are not primarily extraction artefacts
Referential identity 8 duplicate identifiers with different text Identifier-level conflation if the schema uses the nominal clause code alone
Lexical plurality 121 of 204 eligible lemmas polysemous (59.3 percent) Direct term-to-concept mapping risk
Structural signalling 4,556 trigrams; class-conditioned dominant trigram and POS patterns Clause-role confusion if context is ignored
Representational carry-over Ontology identifier coverage 30 of 603; modal coverage 9 of 165 Under-representation in downstream artefacts
Cross-channel fragmentation 75.7 percent of figure design requirements text-absent; 86.4 percent of text design requirements figure-absent A schema on one channel alone misses most of the other channel’s content

Six modal operators across four force categories; 164 modal clauses of 611 total (modal rate 0.2684).

Modal Operator Force Category Corpus Frequency
shall obligatory 140
must obligatory 3
should advisory 4
may permissive 16
can permissive 13
could permissive 4
shall not prohibitive 4

Polysemy confidence tiers (Section 5.14)

Cross-method agreement assigns confidence tiers over the 204 eligible lemmas.

Tier Agreement Interpretation Count in corpus
Strong 3/3 methods agree on polysemous status High confidence that the term carries multiple senses in this corpus 24
Moderate 2/3 methods agree Probable polysemy; one method diverges 97
Weak 1/3 methods flags polysemy Possible polysemy; insufficient cross-method support residual

Design Science Research iteration register (Section 5.8)

Nine documented DSR iterations (0–7b) across four analytical cycles; the design-search history behind the standardisation schema (visualised in the Design Science Research iteration history for the standardisation schema figure).

Iteration Intent Key Change Key Finding Design Decision
0 Corpus integrity Assembled 611 clause records 8 duplicate IDs with divergent text Flag duplicate IDs as a mandatory schema constraint
1 Polysemy baseline Three-method polysemy protocol 121 of 204 eligible lemmas polysemous Schema must carry explicit ambiguity profiles
2 Schema design Five-layer contract specification All five failure modes addressed by design Schema accepted as decision-complete
3 Reproducibility Full rerun on a fresh environment 5-lemma drift after model correction Environment metadata made mandatory; drift within bounds
4 Multi-angle deepening Five new analytical angles Residuals tail-concentrated; ambiguity role-dependent Claim refined to governed ambiguity management
5 Deontic force Modal-verb and force-type analysis 164 modal clauses; force is class-conditioned modality_profile field justified
6 Ambiguity-delta Five-dimensional profiler Overall delta 0.2514; design_requirement delta 0.5696 Dual-mode contribution boundary established
7 Figures integration Five-analysis pipeline on 189 figure DRs 75.7 percent of figure DRs have no text equivalent Two-channel architecture confirmed necessary
7b Figures parity Deontic, delta, and cross-validation on figures Figures delta 0.9148; channels complementary Asymmetric completeness verified