Contrastive learning helps computers find and name event types automatically

CLUE: A Contrastive Learning Framework for Unsupervised Event Type Induction and Annotation

Computation and Language

Summary

Defining types of events in data usually needs experts who spend a lot of time naming and sorting events, which can be expensive and biased. The authors created a method called CLUE that teaches computers to group and name events on their own without needing human guidance. It uses a way to understand verbs and their meanings together, then groups events by how similar they are. This helps make event type definitions faster and less reliant on human work.

What this means in practice

  • For information extraction teams: Automatically group and label event types in datasets without manual rule creation, speeding up event annotation workflows.
  • For text analytics developers: Integrate unsupervised event induction to improve event recognition and organization in large text corpora, reducing reliance on expert input.

Authors

Pratik Maitra, Lauren Kwon, Qi Li

Abstract

Event ontology is the foundational framework for defining events in a dataset. Event type induction generally involves domain experts who define guidelines for creating event types and their names based on the dataset's domain and scope. However, this process is resource-intensive in terms of annotation costs and is prone to human biases. To alleviate the need for manual event ontology creation, we propose the CLUE framework for event type induction, naming, and annotation. Our main contributions in this work are: 1) designing a novel joint embedding representation for expressing both the verb-centric view and the semantic view of events; 2) event type induction using contrastive loss guided clustering; 3) event naming and refinement. Experiments demonstrate the effectiveness of our method. Our code is available at https://github.com/PratikMaitra/CLUE.