Imperial College London

Data scientists win Outstanding Paper Award for research on knowledge graphs

by

Knowledge graphs

A team from the Data Science Institute have won an award for their research on knowledge graphs, presented at the NeurIPS ENLSP workshop.

Researchers from Imperial’s Data Science Institute including Weihang Zhang, Dr Ovidiu Serban, Jiahao Sun and Professor Yike Guo recently won an Outstanding Paper Award for their paper entitled ‘Collective Knowledge Graph Completion with Mutual Knowledge Distillation’. Knowledge graphs are collections of interlinked descriptions of concepts, entities, relationships and events which put data into context, and are used in a range of applications including data governance, fraud detection, personalised recommendations, chatbots, search tools and other intelligent systems.

The award was presented in the Graph for Natural Language Processing track at the NeurIPS-2022 Efficient Natural Language and Speech Processing Workshop on 2 December 2022 in New Orleans.

The workshop focused on the fundamental challenges to make natural language and speech processing models more efficient in terms of data, model, training and interference. Natural language is a subfield of linguistics, computer science and artificial intelligence that analyses the interactions between computers and human language.

In their winning paper, the team proposed a novel method to address the resource imbalance problem between knowledge graphs of different languages.

According to lead author Weihang Zhang: “The results on DBP5L dataset have shown that all multilingual knowledge graphs can benefit from the collective knowledge transfer in this method, resulting in the state of the art performance on multilingual knowledge graph competition task.”

Many congratulations to the team who will be publishing a longer version of the paper later next year.

Reporter

Gemma Ralton

Gemma Ralton
Faculty of Engineering

Click to expand or contract

Contact details

Email: gemma.ralton@imperial.ac.uk

Show all stories by this author

Tags:

Big-data, Global-challenges-Data, Artificial-intelligence
See more tags