Recognition of Named-Event Passages in News Articles

TL;DR

This paper introduces a method to identify specific passages about named events in news articles, extending named entity recognition to event detection, and discusses data collection challenges and solutions.

Contribution

It proposes a novel approach for locating event-specific passages and a practical method for creating gold standard datasets using crowdsourcing.

Findings

Preliminary evaluation results show promise.

Crowdsourcing via Mechanical Turk is effective for data collection.

The method improves event passage identification accuracy.

Abstract

We extend the concept of Named Entities to Named Events - commonly occurring events such as battles and earthquakes. We propose a method for finding specific passages in news articles that contain information about such events and report our preliminary evaluation results. Collecting "Gold Standard" data presents many problems, both practical and conceptual. We present a method for obtaining such data using the Amazon Mechanical Turk service.