Knowledge‐enriched joint‐learning model for implicit emotion cause extraction

Emotion cause extraction (ECE) task that aims at extracting potential trigger events of certain emotions has attracted extensive attention recently. However, current work neglects the implicit emotion expressed without any explicit emotional keywords, which appears more frequently in application scenarios. The lack information makes it extremely hard to extract causes only with local context. Moreover, an entire event is usually across multiple clauses, while existing merely extracts clause level and cannot effectively capture complete information. To address these issues, are first redefined tuple a span-based tuple-level algorithm proposed from different clauses. Based on it, corpus for tries constructed. authors propose knowledge-enriched joint-learning model recognition tasks (KJ-IECE), leverages commonsense knowledge ConceptNet NRC_VAD better connections between corresponding events. Experiments both datasets demonstrate effectiveness model.