Operational Substrate for Personal-Scale Multi-Modal Knowledge Graphs — Open Teaser

EN: This is an open teaser for an independent working paper by Andreas Ehstand (Independent Researcher). Methodological architecture specification, NOT empirical finding and NOT clinical or therapeutic claim. The full paper is in a restricted-access Zenodo companion record (cross-referenced via Related Identifiers); access by request. The substrate combines four properties in a specific configuration: (i) multi-modal ingestion of personal data streams including visual, audio, biometric, and physiological-sensor signals, with provisions for prospective non-invasive neural-signal integration; (ii) subject-authored bidirectional semantic anchoring against a self-validated term ontology; (iii) local hardware sovereignty with encryption at rest and optional air-gap provisions for the most sensitive streams; (iv) longitudinal lifetime-horizon design with explicit substrate-distribution provisions. The substrate operates downstream from a separately-anchored methodology layer (DOI 10.5281/zenodo.20309991, restricted, published 20 May 2026) that formalizes the bidirectional lexical sondierung methodology applied here to non-textual modalities. To the verified knowledge of a May 2026 literature and product survey, the combination of all four constituent properties in a single subject-authored program operated on locally-sovereign infrastructure is not documented in publicly identifiable adjacent work. The differentiation claim is framed conservatively and against publicly identifiable adjacent work only. Living Document: This working teaser is a Living Document. Revisions are tracked. Refinement window per Zenodo policy applies. AI-Assistance Disclosure: Produced with AI assistance (Claude Opus 4.7 by Anthropic) under full human editorial review. EU AI Act Article 50 compliant. DE: Dies ist eine Open-Access-Vorschau einer operativen Substrat-Architektur für mehrmodale persönlich-skalierte Wissensgraph-Konstruktion an einem einzelnen Subjekt. Die vollständige technische Spezifikation ist zugriffsbeschränkt in einem separaten Datensatz verankert; Zugang erfordert autorisierte akademische Korrespondenz. Lizenz CC BY-NC-ND 4.0.