An Evaluation Framework for Mapping News Headlines to Event Classes in a Knowledge Graph

TL;DR

This paper introduces a benchmark dataset and evaluation resources for mapping news headlines to event classes in a knowledge graph, comparing entity linking and zero-shot classification methods.

Contribution

It provides a new dataset and evaluation framework for mapping news headlines to event classes, and assesses various unsupervised methods including entity linking and zero-shot classification.

Findings

Entity linking systems show varying performance on the task.

Pre-trained NLI and language models offer promising results.

Lessons learned guide future research directions.

Abstract

Mapping ongoing news headlines to event-related classes in a rich knowledge base can be an important component in a knowledge-based event analysis and forecasting solution. In this paper, we present a methodology for creating a benchmark dataset of news headlines mapped to event classes in Wikidata, and resources for the evaluation of methods that perform the mapping. We use the dataset to study two classes of unsupervised methods for this task: 1) adaptations of classic entity linking methods, and 2) methods that treat the problem as a zero-shot text classification problem. For the first approach, we evaluate off-the-shelf entity linking systems. For the second approach, we explore a) pre-trained natural language inference (NLI) models, and b) pre-trained large generative language models. We present the results of our evaluation, lessons learned, and directions for future work. The dataset and scripts for evaluation are made publicly available.