Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Deontic Knowledge Graphs for Privacy Compliance in Multimodal Disaster Data Sharing

Kelvin Uzoma Echenim, Karuna Pande Joshi

TL;DR

The paper tackles the privacy-compliant sharing of multimodal disaster data under urgent response needs by moving beyond binary access control to a deontic knowledge-graph framework. It combines a Disaster Management KG (DKG) with a Policy Knowledge Graph (PKG) that encodes FEMA/DHS privacy rules using the IoT-Reg deontic ontology, enabling a release decision function with outcomes Allow, Block, or Allow-with-Transform, and binds obligations to executable transforms verified via provenance. Key contributions include the design of the Disaster Management KG, extension of IoT-Reg with disaster-specific policy concepts, the formal policy-aware decision framework with safety properties, an end-to-end enforcement layer with a transform pipeline and incident logging, and an analytical monitoring layer for federated policy-data queries; evaluation on a 5.1M-triple DKG with 316K images shows exact-match policy decisions and sub-second per-decision latency, with scalable federated queries. The approach supports rapid, auditable, and audience-aware data sharing in disaster response and generalizes to other domains requiring policy-compliant handling of sensitive multimodal data. Practical impact includes automated obfuscation/encryption transforms, provenance-backed artifact derivation, and DHS-aligned incident auditing, enabling compliant and timely situational awareness.

Abstract

Disaster response requires sharing heterogeneous artifacts, from tabular assistance records to UAS imagery, under overlapping privacy mandates. Operational systems often reduce compliance to binary access control, which is brittle in time-critical workflows. We present a novel deontic knowledge graph-based framework that integrates a Disaster Management Knowledge Graph (DKG) with a Policy Knowledge Graph (PKG) derived from IoT-Reg and FEMA/DHS privacy drivers. Our release decision function supports three outcomes: Allow, Block, and Allow-with-Transform. The latter binds obligations to transforms and verifies post-transform compliance via provenance-linked derived artifacts; blocked requests are logged as semantic privacy incidents. Evaluation on a 5.1M-triple DKG with 316K images shows exact-match decision correctness, sub-second per-decision latency, and interactive query performance across both single-graph and federated workloads.

Deontic Knowledge Graphs for Privacy Compliance in Multimodal Disaster Data Sharing

TL;DR

The paper tackles the privacy-compliant sharing of multimodal disaster data under urgent response needs by moving beyond binary access control to a deontic knowledge-graph framework. It combines a Disaster Management KG (DKG) with a Policy Knowledge Graph (PKG) that encodes FEMA/DHS privacy rules using the IoT-Reg deontic ontology, enabling a release decision function with outcomes Allow, Block, or Allow-with-Transform, and binds obligations to executable transforms verified via provenance. Key contributions include the design of the Disaster Management KG, extension of IoT-Reg with disaster-specific policy concepts, the formal policy-aware decision framework with safety properties, an end-to-end enforcement layer with a transform pipeline and incident logging, and an analytical monitoring layer for federated policy-data queries; evaluation on a 5.1M-triple DKG with 316K images shows exact-match policy decisions and sub-second per-decision latency, with scalable federated queries. The approach supports rapid, auditable, and audience-aware data sharing in disaster response and generalizes to other domains requiring policy-compliant handling of sensitive multimodal data. Practical impact includes automated obfuscation/encryption transforms, provenance-backed artifact derivation, and DHS-aligned incident auditing, enabling compliant and timely situational awareness.

Abstract

Disaster response requires sharing heterogeneous artifacts, from tabular assistance records to UAS imagery, under overlapping privacy mandates. Operational systems often reduce compliance to binary access control, which is brittle in time-critical workflows. We present a novel deontic knowledge graph-based framework that integrates a Disaster Management Knowledge Graph (DKG) with a Policy Knowledge Graph (PKG) derived from IoT-Reg and FEMA/DHS privacy drivers. Our release decision function supports three outcomes: Allow, Block, and Allow-with-Transform. The latter binds obligations to transforms and verifies post-transform compliance via provenance-linked derived artifacts; blocked requests are logged as semantic privacy incidents. Evaluation on a 5.1M-triple DKG with 316K images shows exact-match decision correctness, sub-second per-decision latency, and interactive query performance across both single-graph and federated workloads.
Paper Structure (64 sections, 1 equation, 3 figures, 4 tables, 2 algorithms)

This paper contains 64 sections, 1 equation, 3 figures, 4 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Policy-Aware Disaster Data Sharing
  3. Challenges of Multimodal Privacy Compliance
  4. Heterogeneous Data Modalities.
  5. Contextual Integrity and Audience-Dependent Policies.
  6. Dynamic Compliance States.
  7. Regulatory Interplay.
  8. Our Contribution
  9. Background and Problem Setting
  10. Deontic Logic and the IoT-Reg Framework
  11. IoT-Reg as a Deontic Ontology for IoT Privacy.
  12. Regulatory Drivers
  13. NISTIR 8228: The IoT Privacy Baseline
  14. FEMA SORN DHS/FEMA-008: Structured PII Limits
  15. FEMA UAS PIA-055: Unstructured Imagery Mandates
  16. ...and 49 more sections

Figures (3)

  • Figure 1: The Disaster Management Ontology.
  • Figure 2: Knowledge graph population pipeline showing external data sources, the DKG Builder processing component, and the resulting SPARQL endpoint hosting both the Disaster KG and Policy KG, with policy rules derived from FEMA/DHS regulatory documents.
  • Figure 3: End-to-end architecture of the privacy-aware disaster data sharing framework, showing the flow from release requests through policy evaluation, transform execution, and compliance verification, with the Disaster KG and Policy KG as dual knowledge sources.