Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Pattern-Aware and Noise-Resilient Embedding Models

: Nayyeri, M.; Vahdati, S.; Sallinger, E.; Alam, M.M.; Yazdi, H.S.; Lehmann, J.


Hiemstra, D.:
Advances in Information Retrieval. 43rd European Conference on IR Research, ECIR 2021. Proceedings. Pt.I : Virtual Event, March 28 - April 1, 2021
Cham: Springer Nature, 2021 (Lecture Notes in Computer Science 12656)
ISBN: 978-3-030-72112-1 (Print)
ISBN: 978-3-030-72113-8 (Online)
ISBN: 978-3-030-72114-5
European Conference on IR Research (ECIR) <43, 2021, Online>
Fraunhofer IAIS ()

Knowledge Graph Embeddings (KGE) have become an important area of Information Retrieval (IR), in particular as they provide one of the state-of-the-art methods for Link Prediction. Recent work in the area of KGEs has shown the importance of relational patterns, i.e., logical formulas, to improve the learning process of KGE models significantly. In separate work, the role of noise in many knowledge discovery and IR settings has been studied, including the KGE setting. So far, very few papers have investigated the KGE setting considering both relational patterns and noise. Not considering both together can lead to problems in the performance of KGE models. We investigate the effect of noise in the presence of patterns. We show that by introducing a new loss function that is both pattern-aware and noise-resilient, significant performance issues can be solved. The proposed loss function is model-independent which could be applied in combination with different models. We pro vide an experimental evaluation both on synthetic and real-world cases.