Event-Driven News Stream Clustering using Entity-Aware Contextual Embeddings

Kailash Karthik Saravanakumar, Miguel Ballesteros, Muthu Kumar Chandrasekaran, Kathleen McKeown


Abstract
We propose a method for online news stream clustering that is a variant of the non-parametric streaming K-means algorithm. Our model uses a combination of sparse and dense document representations, aggregates document-cluster similarity along these multiple representations and makes the clustering decision using a neural classifier. The weighted document-cluster similarity model is learned using a novel adaptation of the triplet loss into a linear classification objective. We show that the use of a suitable fine-tuning objective and external knowledge in pre-trained transformer models yields significant improvements in the effectiveness of contextual embeddings for clustering. Our model achieves a new state-of-the-art on a standard stream clustering dataset of English documents.
Anthology ID:
2021.eacl-main.198
Volume:
Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume
Month:
April
Year:
2021
Address:
Online
Venue:
EACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2330–2340
Language:
URL:
https://aclanthology.org/2021.eacl-main.198
DOI:
10.18653/v1/2021.eacl-main.198
Bibkey:
Cite (ACL):
Kailash Karthik Saravanakumar, Miguel Ballesteros, Muthu Kumar Chandrasekaran, and Kathleen McKeown. 2021. Event-Driven News Stream Clustering using Entity-Aware Contextual Embeddings. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 2330–2340, Online. Association for Computational Linguistics.
Cite (Informal):
Event-Driven News Stream Clustering using Entity-Aware Contextual Embeddings (Saravanakumar et al., EACL 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.eacl-main.198.pdf