ICDM 2020 Knowledge Graph Contest: Consumer Event-Cause Extraction

TL;DR

This paper tackles Consumer Event-Cause Extraction (CECE) in the ICDM 2020 Knowledge Graph Contest by proposing an end-to-end sequence tagging framework that jointly extracts multiple consumer event types and their causes for a given brand/product. The model combines a BERT encoder with a CRF-based sequence tagging decoder, moving beyond traditional MRC-style approaches that treat event types and causes separately. An ensemble via BERT stacking further boosts performance, with results showing clear gains over baselines and leading to top leaderboard placements. The approach enables robust end-to-end extraction suitable for business applications like advertising and social listening, where explicit event-cause relationships are valuable.

Abstract

Consumer Event-Cause Extraction, the task aimed at extracting the potential causes behind certain events in the text, has gained much attention in recent years due to its wide applications. The ICDM 2020 conference sets up an evaluation competition that aims to extract events and the causes of the extracted events with a specified subject (a brand or product). In this task, we mainly focus on how to construct an end-to-end model, and extract multiple event types and event-causes simultaneously. To this end, we introduce a fresh perspective to revisit the relational event-cause extraction task and propose a novel sequence tagging framework, instead of extracting event types and events-causes separately. Experiments show our framework outperforms baseline methods even when its encoder module uses an initialized pre-trained BERT encoder, showing the power of the new tagging framework. In this competition, our team achieved 1st place in the first stage leaderboard, and 3rd place in the final stage leaderboard.

Figures (2)

• Figure 1: Overview of our model, our proposed approach for consumer event-cause extraction.
• Figure 2: Ensemble strategy based on BERT stacking and weighting operation.