AI Identity Diagnostic: A Verified Person Identity Consolidation Case Study
Year: 2026
Type: article
Abstract: This technical report documents a real-world production case study using Sydney Business Web's AI Identity Diagnostic to identify, interpret and remediate fragmented Person identities within a whole-site structured-data entity graph. An initial Schema Gorilla analysis of 274 website pages identified two additional Keith Rowley Person representations that were not connected to the established canonical Person identity. These unresolved fragments produced four Person clusters and an ATTENTION verdict for identity consolidation. Following targeted corrections to the affected author relationships, a fresh Discovery and Schema Gorilla analysis examined the same 274 pages. The verification recorded: Person clusters: Reduced from 4 to 2. Total entity clusters: Reduced from 28 to 26. Pages analysed: 274 before and after remediation. Person identity consolidation: Changed from ATTENTION to PASS. The report explains the diagnostic methodology, graph interpretation, canonical identifier consolidation, engineering remediation and independent fresh-run verification. It distinguishes Schema Gorilla's structural analysis from AI Identity Diagnostic's interpretation layer and AI Observatory's separate retrieval measurements. The findings demonstrate a repeatable, evidence-based method for identifying and correcting machine-readable business identity fragmentation. The PASS verdict applies specifically to the tested Person identity consolidation check and does not establish improved AI search rankings, citations, external corroboration or machine understanding. The investigation is a developer-operated, single-site case study. Its reported results have not been independently audited. Author: Keith Rowley, Sydney Business Web. Publication: Engineering Research Report 09, Version 1.0.0, 14 November 2026.
Author: Rowley, Keith
Language: English
Topic: Data Quality and Management
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7228050282
