TransO: a knowledge-driven representation learning method with ontology information constraints

Representation learning techniques for knowledge graphs (KGs) are crucial for constructing knowledge-driven decisions in complex network data application scenarios. Most existing methods focus mainly on structured information, ignoring the important value of rich ontology information constraints and complements, however, ontology information is the key for building knowledge-driven decision-making processes. In this paper, we propose a novel ontology information constrained knowledge representation learning model, TransO, which can efficiently model relations explicitly and seamlessly incorporate rich ontology information to improve model performance and maintain low model complexity. Moreover, specific constraint strategies are proposed for entity types, relations, and hierarchical information to effectively implement reasoning and completion of KGs and construct knowledge-driven decisions that are more consistent with the logic of human knowledge in complex network applications. The experimental tasks of link prediction and triple classification are performed on two public datasets. The experimental results demonstrate the effectiveness of our proposed method with better performance than state-of-the-art methods.