Conversation Derailment Forecasting with Graph Convolutional Networks
This work addresses the issue of toxic communication in online conversations, enabling proactive moderation, but it is incremental with small gains over existing methods.
The paper tackles the problem of forecasting conversation derailment in online dialogues by proposing a graph convolutional neural network model that incorporates user dynamics and public perception, achieving performance improvements of 1.5% and 1.7% over state-of-the-art models on benchmark datasets.
Online conversations are particularly susceptible to derailment, which can manifest itself in the form of toxic communication patterns like disrespectful comments or verbal abuse. Forecasting conversation derailment predicts signs of derailment in advance enabling proactive moderation of conversations. Current state-of-the-art approaches to address this problem rely on sequence models that treat dialogues as text streams. We propose a novel model based on a graph convolutional neural network that considers dialogue user dynamics and the influence of public perception on conversation utterances. Through empirical evaluation, we show that our model effectively captures conversation dynamics and outperforms the state-of-the-art models on the CGA and CMV benchmark datasets by 1.5\% and 1.7\%, respectively.