Part of human’s understanding of the world can be stored as various kinds of graph-structured representations, such as knowledge graphs, lexical graphs, product graphs, social networks, and biological networks. The essence of such structured knowledge representation is to provide a way of modeling the diverse types of relations for entities, concepts, and human language components, as well as interactions between molecules and biomolecules in nature. Accordingly, those sources of actionable knowledge are key to the support of many knowledge-aware intelligent systems for natural language understanding, e-commerce recommendation, social media analysis, and in silico research for biology and medicine. Particularly in this era of big data where intelligent systems are usually data-driven and learning-based, structured knowledge often supports the systems with stronger reasoning abilities and better robustness against reporting bias.

This article collection targets cutting edge research efforts that address three key problems for representation and learning of structured knowledge: (i) Knowledge acquisition: data-driven methods for acquiring structured knowledge from unstructured data, aligning and synchronizing the knowledge among different sources of data; (ii) Knowledge-based inference: inference of proximity, types and relations, as well as prediction and verification of missing knowledge; (iii) Knowledge-based application: successful ways of applying structured knowledge representations in downstream applications.

Detailed topics of interest include but not limited to the following:
–Representation and resources of structured knowledge: techniques for the semantic web; knowledge graph construction; domain-specific or application-driven knowledge graph; symbolic or distributional knowledge representations; causal graph inference; physical graph construction.
–Learning and inference of structured knowledge: fundamental research of representation learning, relational learning, and reasoning with constraints of relations; techniques for knowledge graph completion, entity type inference, anomaly detection.
–Integration and synchronization of structured knowledge: fundamental research of entity resolution, knowledge graph alignment and network matching; techniques for data integration between structured data, or grounding of unstructured data to structured data.
–Structured knowledge discovery in data analytics: information extraction including entity typing, named entity recognition, entity linking, relation extraction, event detection and extraction; ontology induction; scene graph generation; visual relation detection.
–Structured knowledge for natural language processing: Knowledge base question answering, knowledge-enhanced chatbot; zero-shot classification with knowledge; few-shot learning with knowledge; improving the models’ generalization ability with knowledge; knowledge-enhanced information extraction; defending adversarial attack with knowledge.
–Structured knowledge in computational biology and medicine research: drug discovery; molecule property classification; chemical reaction prediction; molecule graph generation; protein-protein-interaction/drug-drug-interaction/drug-target-interaction prediction.

Keywords: Representation Learning, Knowledge Graph, Heterogeneous Information Network, Natural Language Understanding


Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.

About Frontiers Research Topics
With their unique mixes of varied contributions from Original Research to Review Articles, Research Topics unify the most influential researchers, the latest key findings and historical advances in a hot research area! Find out more on how to host your own Frontiers Research Topic or contribute to one as an author.