4th NLP4KGC

International Workshop on Natural Language Processing  for Knowledge Graph Construction



Sydney, Australia

April 2025

Supported by

4th International Workshop on Natural Language Processing for Knowledge Graph Construction in conjunction with The Web Conference 2025.


Call for Paper


This is the fourth edition of the workshop “Natural Language Processing for Knowledge Graph Creation – NLP4KGC” 2025.

The workshop focuses on methods and approaches of knowledge and data extraction from text, as well as theoretical and practical aspects of using semantics and Natural Language Processing for Knowledge Graph creation, and the use of such knowledge graphs for Graph Neural Network (GNN) tasks.

Knowledge Graphs (KG) are emerging and becoming increasingly popular. Building a domain knowledge graph from a large amount of text requires a tremendous amount of work, including entity recognition, entity disambiguation and relationship extraction. Because of this, there has been more work done on automated ways to generate knowledge graphs from text. Recent efforts in NLP development have shown that semantic deep neural networks can learn the complex syntactic and semantics of the natural language and thus, give more potential for automation even in the most complex domains i.e., legal documents. New approaches to Knowledge Graph development use a combination of extraction methods and state-of-the-art Natural Language Processing (NLP) techniques. Recently, considerable literature in this space has centered around the use of Graph Neural Networks (GNNs) to learn powerful embeddings which leverage topological structures in the KGs. Despite the successes this existing research has achieved, deep learning on graphs for NLP still faces many challenges. Presentation of challenges faced in specific domains such as Science, Sustainability are also are welcomed. For instance in the sciences, the production of resources (e.g., publications) is growing at a rate that outstrips an individual’s capacity to discovery and utilize them. As a result, alongside manual methods to structure and make discoverable that knowledge, automation is vital. We welcome applications that analyze and synthesize large volumes of scientific information (especially scientific literature, notes documents, and dataset metadata) and can use that information to create ontologies and semantic representations that better organize them. We also welcome contributions that address automation components of maintenance of ontologies and knowledge graphs. 


This workshop invites contributions on methods and approaches of knowledge and data extraction from text, as well as theoretical and practical aspects of using semantic deep NLP for KG creation and the use of such KG for Graph Neural Network (GNN) tasks.


Important Dates

All submission deadlines are end-of-day in the Anywhere on Earth (AoE) time zone. 

Topics

Here are the major topics but not limited:

– Entity extraction and resolution from text for knowledge graph construction.

– Text entity relationship extraction and linking for knowledge graph construction.

– Entity disambiguation.

– Large language models.

– Multilingual/language-specific information retrieval for knowledge graph construction.

– Ontology learning from text.

– Neural network architectures for knowledge graph construction from text.

– Dynamic knowledge graph building from text/text streams.

– Explainable AI for knowledge graph construction from text.

– GNNs for NLP (graph construction, graph representation learning).

– Use of LLM agents, tools, and Retrieval Augmented Generation, GraphRAG.

– Knowledge Graphs for Bias Mitigation.

– Interpretability and Explainability Knowledge Graph driven assessment 

– Responsible AI frameworks.

– Automatic annotation of web pages with schema.org.

– Distant supervision with web-scale data such as Web Data Commons.

– Semantic Web, Web Data, and Linked Data


Submission Guidelines

We invite full research papers, negative results, position papers, dataset and system demo papers. Papers must be written in English, in double-column format, and must adhere to the ACM template and formatting (also available in Overleaf). The page limit for the full research papers, negative results and dataset papers is 8 pages excluding references and for the short papers and demos it is 4 pages excluding references.  

Submissions must be original and should not have been published previously or be under consideration for publication while being evaluated for this workshop. Submissions will be evaluated by the program committee based on the quality of the work and its fit to the workshop themes. All submissions are double-blind and a high-resolution PDF of the paper should be uploaded to the EasyChair submission site before the paper submission deadline.  The accepted papers will be presented at the NLP4KGC workshop integrated with the conference, and they will be published at ACM Digital Library. 

If you have any questions, please contact us via email: sima.iranmanesh@bcu.ac.uk

Submission Site: Submissions must be made through the EasyChair system.