Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Online Detection of Anomalies in Temporal Knowledge Graphs with Interpretability

Jiasheng Zhang, Rex Ying, Jie Shao

TL;DR

AnoT, an efficient TKG summarization method tailored for interpretable online anomaly detection in TKGs, is introduced, encompassing a detector for offline TKG summarization and online scoring, an updater for real-time rule graph updates based on emerging knowledge, and a monitor for estimating the approximation error of the rule graph.

Abstract

Temporal knowledge graphs (TKGs) are valuable resources for capturing evolving relationships among entities, yet they are often plagued by noise, necessitating robust anomaly detection mechanisms. Existing dynamic graph anomaly detection approaches struggle to capture the rich semantics introduced by node and edge categories within TKGs, while TKG embedding methods lack interpretability, undermining the credibility of anomaly detection. Moreover, these methods falter in adapting to pattern changes and semantic drifts resulting from knowledge updates. To tackle these challenges, we introduce AnoT, an efficient TKG summarization method tailored for interpretable online anomaly detection in TKGs. AnoT begins by summarizing a TKG into a novel rule graph, enabling flexible inference of complex patterns in TKGs. When new knowledge emerges, AnoT maps it onto a node in the rule graph and traverses the rule graph recursively to derive the anomaly score of the knowledge. The traversal yields reachable nodes that furnish interpretable evidence for the validity or the anomalous of the new knowledge. Overall, AnoT embodies a detector-updater-monitor architecture, encompassing a detector for offline TKG summarization and online scoring, an updater for real-time rule graph updates based on emerging knowledge, and a monitor for estimating the approximation error of the rule graph. Experimental results on four real-world datasets demonstrate that AnoT surpasses existing methods significantly in terms of accuracy and interoperability. All of the raw datasets and the implementation of AnoT are provided in https://github.com/zjs123/ANoT.

Online Detection of Anomalies in Temporal Knowledge Graphs with Interpretability

TL;DR

AnoT, an efficient TKG summarization method tailored for interpretable online anomaly detection in TKGs, is introduced, encompassing a detector for offline TKG summarization and online scoring, an updater for real-time rule graph updates based on emerging knowledge, and a monitor for estimating the approximation error of the rule graph.

Abstract

Temporal knowledge graphs (TKGs) are valuable resources for capturing evolving relationships among entities, yet they are often plagued by noise, necessitating robust anomaly detection mechanisms. Existing dynamic graph anomaly detection approaches struggle to capture the rich semantics introduced by node and edge categories within TKGs, while TKG embedding methods lack interpretability, undermining the credibility of anomaly detection. Moreover, these methods falter in adapting to pattern changes and semantic drifts resulting from knowledge updates. To tackle these challenges, we introduce AnoT, an efficient TKG summarization method tailored for interpretable online anomaly detection in TKGs. AnoT begins by summarizing a TKG into a novel rule graph, enabling flexible inference of complex patterns in TKGs. When new knowledge emerges, AnoT maps it onto a node in the rule graph and traverses the rule graph recursively to derive the anomaly score of the knowledge. The traversal yields reachable nodes that furnish interpretable evidence for the validity or the anomalous of the new knowledge. Overall, AnoT embodies a detector-updater-monitor architecture, encompassing a detector for offline TKG summarization and online scoring, an updater for real-time rule graph updates based on emerging knowledge, and a monitor for estimating the approximation error of the rule graph. Experimental results on four real-world datasets demonstrate that AnoT surpasses existing methods significantly in terms of accuracy and interoperability. All of the raw datasets and the implementation of AnoT are provided in https://github.com/zjs123/ANoT.
Paper Structure (41 sections, 11 equations, 10 figures, 7 tables, 3 algorithms)

This paper contains 41 sections, 11 equations, 10 figures, 7 tables, 3 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. Dynamic graph anomaly detection.
  4. TKG embedding.
  5. Graph summarization.
  6. Preliminaries
  7. Temporal Knowledge Graph
  8. Anomalies in TKGs
  9. Conceptual Errors.
  10. Time Errors.
  11. Missing Errors.
  12. Minimum Description Length Principle
  13. Summarization of A TKG
  14. Atomic Rules.
  15. Rule Graph.
  16. ...and 26 more sections

Figures (10)

  • Figure 1: An illustration of different types of anomalies in TKGs and how anomalies relate to preserved knowledge.
  • Figure 2: Conceptual illustration of the proposed AnoT framework.
  • Figure 3: Rule graph construction, which contains: 1. Category function construction, 2. Generating candidate atomic rules, 3. Generating candidate rule edges, 4. Ranking and selecting.
  • Figure 4: Conceptual illustrations of the scoring process and the updater module.
  • Figure 5: Performance of ANoT and RE-GCN when different proportions of offline preserved knowledge are used to construct the optimal rule graph. Concept, time, and missing respectively refer to conceptual errors, time errors, and missing errors.
  • ...and 5 more figures

Theorems & Definitions (2)

  • definition 1: Inductive anomaly detection in TKGs
  • definition 2: Inductive TKG summarization with MDL