We are pleased to announce that the doctoral thesis "Semantic and Efficient Symbolic Learning over Knowledge Graphs" by Disha Purohit is now available in our institutional repository.
This research addresses a critical limitation of modern Artificial Intelligence: the incomplete trustworthiness of Knowledge Graphs (KGs). These KGs, essential for AI systems, often contain noisy or incomplete data.
Key Contribution: Purohit introduces a neuro-symbolic framework that combines the power of neural networks with the transparency and logical consistency of symbolic reasoning. This approach ensures that the inferred knowledge is not only statistically plausible but also semantically sound and verifiable against domain ontologies.
The work establishes a new paradigm for Knowledge Graph Completion (KGC), leading to more reliable and explainable AI systems.
