Enhancing Knowledge Graph Embedding with Uncertainty Modeling using Fuzzy Logic

October 16th, 2023

Thesis Type
  • Master
Presentation room
Seminar room I5 6202
Stefan Decker
Gerhard Lakemeyer
Yongli Mou

Knowledge graphs have emerged as powerful tools for organizing and querying structured information. Knowledge graph embedding, a fundamental task in knowledge graph analysis, aims to learn continuous vector representations for entities and relations in a knowledge graph, enabling the application of various machine learning and knowledge discovery techniques. However, real-world knowledge graphs often contain noisy, incomplete, or uncertain information, which poses significant challenges for traditional deterministic knowledge graphs and knowledge graph embedding methods.

This thesis focuses on the intersection of knowledge graph embedding and uncertainty modeling, with a specific emphasis on incorporating Fuzzy Logic (FL), more specifically Probabilistic Soft Logic (PSL), into the embedding process. PSL is a framework that blends probabilistic reasoning with first-order logic, offering a flexible and intuitive way to represent and reason about uncertainty in knowledge graphs.


  • [1] Chen X, Chen M, Shi W, Sun Y, Zaniolo C. Embedding uncertain knowledge graphs. InProceedings of the AAAI conference on artificial intelligence 2019 Jul 17 (Vol. 33, No. 01, pp. 3363-3370).
  • [2] Cao Z, Xu Q, Yang Z, Cao X, Huang Q. Geometry interaction knowledge graph embeddings. InProceedings of the AAAI Conference on Artificial Intelligence 2022 Jun 28 (Vol. 36, No. 5, pp. 5521-5529).
  • [3] H├íjek P. Metamathematics of fuzzy logic. Springer Science & Business Media; 2013 Dec 1.


Deep knowledge in mathematical logic

Deep knowledge in machine learning