Abstract
Most of the existing knowledge representation learning methods project the entities and relations represented by symbols in the knowledge graph into the low-dimensional vector space from the perspective of the structure and semantics of triples, and express the complex relations between entities and relations with dense low-dimensional vectors. However, triples in the knowledge graph not only contain relation triples, but also contain a large number of attribute triples. Existing knowledge representation methods often confuse these two kinds of triples and pay little attention to the semantic information contained in attributes and attribute values. In this paper, a novel representation learning method which makes use of the attribute information of entities is proposed. Specifically, deep convolutional neural network model is used to encode attribute information of entities, and both attribute information and triple structure information are utilized to learn knowledge representation, and then generate attribute-based representation of entities. The knowledge graph completion task was used to evaluate this method, and the experimental results on open data sets FB15K and FB24k showed that the attribute-embodied knowledge representation learning model outperforms the other baselines.
Highlights
In a narrow sense, knowledge graph was first proposed by Google in 2012, and was used by Internet companies to organize network data from the semantic perspective, so as to provide a large knowledge base of intelligent search services
Experimental results show that in the FB15K data set with more relational triples, the effect of each knowledge representation method is not much different, while in the FB24K data set with more attribute triples, it is obvious that the Attribute-embodied Knowledge Representation Learning (AKRL) model has more advantages than the traditional TransE and TransR models, as well as the latest Description-Embodied Knowledge Representation Learning (DKRL) and KR-EAR models
If attribute triples are regarded as relation triples in the process of representation learning, the non-entity information in the triples will cause error diffusion in the training process, and lead to a cliff drop in learning effect
Summary
Knowledge graph was first proposed by Google in 2012, and was used by Internet companies to organize network data from the semantic perspective, so as to provide a large knowledge base of intelligent search services. Companies from all walks of life are exploring the establishment of knowledge graph in vertical fields, so as to enhance the intelligence level of financial, medical, judicial, educational and publishing business. Researchers are working on ways to automate the construction and application of knowledge graphs. The goal of representation learning is to represent the semantic information of an object with dense low-dimensional real value vectors through machine learning and other methods [2]. For the application of artificial intelligence, knowledge can only be invoked after being expressed in an appropriate way. Knowledge representation is at the heart part of knowledge-based AI application
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
