Huiping Cao

Trung Hoang Le, Tran Cao Son, Huiping Cao

Logical rule-based methods offer an interpretable approach to knowledge graph completion by capturing compositional relationships in the form of human-readable inference rules. However, current approaches typically treat logical rules as universal, assigning each rule a fixed confidence score that ignores query-specific context. This is a significant limitation, as a rule's importance can vary depending on the query. To address this, we introduce SLogic (Subgraph-Informed Logical Rule learning), a novel framework that assigns query-dependent scores to logical rules. The core of SLogic is a scoring function that utilizes the subgraph centered on a query's head entity, allowing the significance of each rule to be assessed dynamically. Extensive experiments on benchmark datasets show that by leveraging local subgraph context, SLogic consistently outperforms state-of-the-art baselines, including both embedding-based and rule-based methods.

James Obert, Huiping Cao

Multipath routing is the use of multiple potential paths through a network in order to enhance fault tolerance, optimize bandwidth use, and improve security. Selecting data flow paths based on cost addresses performance issues but ignores security threats. Attackers can disrupt the data flows by attacking the links along the paths. Denial-of-service, remote exploitation, and other such attacks launched on any single link can severely limit throughput. Networks can be secured using a secure quality of service approach in which a sender disperses data along multiple secure paths. In this secure multi-path approach, a portion of the data from the sender is transmitted over each path and the receiver assembles the data fragments that arrive. One of the largest challenges in secure multipath routing is determining the security threat level along each path and providing a commensurate level of encryption along that path. The research presented explores the effects of real-world attack scenarios in systems, and gauges the threat levels along each path. Optimal sampling and compression of network data is provided via compressed sensing. The probability of the presence of specific attack signatures along a network path is determined using machine learning techniques. Using these probabilities, information assurance levels are derived such that security measures along vulnerable paths are increased.