From Discourse to Narrative: Knowledge Projection for Event Relation Extraction

Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, pages 732–742 August 1–6, 2021. ©2021 Association for Computational Linguistics 732 From Discourse to Narrative: Knowledge Projection for Event Relation Extraction Jialong Tang1,3, Hongyu Lin1, Meng Liao4, Yaojie Lu1,3, Xianpei Han1,2, Le Sun1,2,∗, Weijian Xie4, Jin Xu4,∗ 1Chinese Information Processing Laboratory 2State Key Laboratory of Computer Science Institute of Software, Chinese Academy of Sciences, Beijing, China 3University of Chinese Academy of Sciences, Beijing, China 4Data Quality Team, WeChat, Tencent Inc., China {jialong2019,hongyu,xianpei,sunle}@iscas.ac.cn {maricoliao, vikoxie, jinxxu}@tencent.com Abstract Current event-centric knowledge graphs highly rely on explicit connectives to mine relations between events. Unfortunately, due to the spar- sity of connectives, these methods severely undermine the coverage of EventKGs. The lack of high-quality labelled corpora further exacerbates that problem. In this paper, we propose a knowledge projection paradigm for event relation extraction: projecting discourse knowledge to narratives by exploiting the commonalities between them. Specifically, we propose Multi-tier Knowledge Projection Network (MKPNet), which can leverage multi- tier discourse knowledge effectively for event relation extraction.In this way, the labelled data requirement is significantly reduced, and implicit event relations can be effectively ex- tracted. Intrinsic experimental results show that MKPNet achieves the new state-of-the-art per- formance, and extrinsic experimental results verify the value of the extracted event relations. 1 Introduction Event-centric knowledge graphs (EventKGs) model the narratives of the world by represent- ing events and identifying relations between them, which are critical for machine understanding and can benefit many downstream tasks, such as ques- tion answering (Costa et al., 2020), news read- ing (Vossen, 2018), commonsense knowledge ac- quisition (Zhang et al., 2020a) and so on. Recently, semi-automatically constructing Even- tKGs have gained much attention (Tandon et al., 2015; Rospocher et al., 2016; Gottschalk and Demi- dova, 2018; Zhang et al., 2020b). These methods extract event knowledge from massive raw corpora with or without little human intervention, which makes them scalable solutions to build large-scale ∗Corresponding authors. Asynchronous Cause 𝐷�: Tom goes to the restaurant 𝐷�: He orders two hamburgers 𝐷�: because he is so hungry 𝐷� 𝐸�: PER is so hungry Reason (because) Reason Explicit Relations Implicit Relations Precedence Succession 𝐸�: PER orders two hamburgers 𝐸�: PER goes to the restaurant. Knowledge Projection 𝐷� 𝐷� Cause Discourse Narrative Figure 1: The knowledge projection paradigm for event relation extraction. The explicit projection directly projects connectives to event relations, e.g., from “be- cause” to Reason. The implicit projection leverages the discourse knowledge to discover implicit event relations without connectives via MKPNet. EventKGs. Commonly, each node in EventKGs represents an event, and each edge represents a pre- defined relation between an event pair1. Currently, event relations are majorly extracted based on the explicit connectives between them. For example, in Figure 1, a Reason relation is extracted between E2: “PER orders two hamburgers” and E3: “PER is so hungry” using the explicit connective “be- cause” between them. Unfortunately, the connective-based approaches face the critical coverage problem due …

Content truncated to 500 words