Thread Reconstruction in Conversational Data using Neural Coherence Models
This addresses the challenge for users in navigating complex forum discussions, though it appears incremental as it builds on existing coherence modeling approaches.
The paper tackles the problem of automatically identifying the underlying thread structure in forum discussions, which can be intricate and hard to follow, by proposing a neural coherence model that selects the most coherent reconstruction, with preliminary experiments showing promising results and outperforming strong baselines.
Discussion forums are an important source of information. They are often used to answer specific questions a user might have and to discover more about a topic of interest. Discussions in these forums may evolve in intricate ways, making it difficult for users to follow the flow of ideas. We propose a novel approach for automatically identifying the underlying thread structure of a forum discussion. Our approach is based on a neural model that computes coherence scores of possible reconstructions and then selects the highest scoring, i.e., the most coherent one. Preliminary experiments demonstrate promising results outperforming a number of strong baseline methods.