Probabilistic Box Embeddings for Uncertain Knowledge Graph Reasoning
21802Authors
nan
Abstract
AbstractKnowledge bases often consist of facts which are harvested from a variety of sources, many of which are noisy and some of which conflict, resulting in a level of uncertainty for each triple. Knowledge bases are also often incomplete, prompting the use of embedding methods to generalize from known facts, however, existing embedding methods only model triple-level uncertainty, and reasoning results lack global consistency. To address these shortcomings, we propose BEUrRE, a novel uncertain knowledge graph embedding method with calibrated probabilistic semantics. BEUrRE models each entity as a box (i.e. axis-aligned hyperrectangle) and relations between two entities as affine transforms on the head and tail entity boxes. The geometry of the boxes allows for efficient calculation of intersections and volumes, endowing the model with calibrated probabilistic semantics and facilitating the incorporation of relational constraints. Extensive experiments on two benchmark datasets show that BEUrRE consistently outperforms baselines on confidence prediction and fact ranking due to its probabilistic calibration and ability to capture high-order dependencies among facts.
Keywords
Full Content
Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 62–73 June 6–11, 2021. ©2021 Association for Computational Linguistics 62 Event Time Extraction and Propagation via Graph Attention Networks Haoyang Wen1, Yanru Qu1, Heng Ji1, Qiang Ning2∗, Jiawei Han1, Avirup Sil3, Hanghang Tong1, Dan Roth4 1 University of Illinois at Urbana-Champaign 2Amazon 3 IBM Research AI 4University of Pennsylvania {wen17, yanruqu2, hengji, hanj, htong}@illinois.edu qning@amazon.com avi@us.ibm.com danroth@seas.upenn.edu Abstract Grounding events into a precise timeline is im- portant for natural language understanding but has received limited attention in recent work. This problem is challenging due to the inher- ent ambiguity of language and the requirement for information propagation over inter-related events. This paper first formulates this prob- lem based on a 4-tuple temporal representation used in entity slot filling, which allows us to represent fuzzy time spans more conveniently. We then propose a graph attention network- based approach to propagate temporal infor- mation over document-level event graphs con- structed by shared entity arguments and tempo- ral relations. To better evaluate our approach, we present a challenging new benchmark on the ACE2005 corpus, where more than 78% of events do not have time spans mentioned ex- plicitly in their local contexts. The proposed approach yields an absolute gain of 7.0% in match rate over contextualized embedding ap- proaches, and 16.3% higher match rate com- pared to sentence-level manual event time ar- gument annotation.1 1 Introduction Understanding and reasoning about time is a cru- cial component for comprehensive understanding of evolving situations, events, trends and forecast- ing event abstractions for the long-term. Event time extraction is also useful for many downstream Nat- ural Language Processing (NLP) applications such as event timeline generation (Huang and Huang, 2013; Wang et al., 2015; Ge et al., 2015; Steen and Markert, 2019), temporal event tracking and prediction (Ji et al., 2009; Minard et al., 2015), and temporal question answering (Llorens et al., 2015; Meng et al., 2017). *Work done prior to joining Amazon. 1The resource for this paper is available at https://gi thub.com/wenhycs/NAACL2021-Event-Time-Ex traction-and-Propagation-via-Graph-Atten tion-Networks. In order to ground events into a timeline we need to determine the start time and end time of each event as precisely as possible (Reimers et al., 2016). However, the start and end time of an event is often not explicitly expressed in a document. For exam- ple, among 5,271 annotated event mentions in the Automatic Content Extraction (ACE2005) corpus2, only 1,100 of them have explicit time argument annotations. To solve the temporal event ground- ing (TEG) problem, previous efforts focus on its subtasks such as temporal event ordering (Bram- sen et al., 2006; Chambers and Jurafsky, 2008; Yoshikawa et al., 2009; Do et al., 2012; Meng et al., 2017; Meng and Rumshisky, 2018; Ning et al., 2017, 2018, 2019; Han et al., 2019) and duration prediction (Pan et al., 2006, 2011; Vempala et al., 2018; Gusev et al., 2011; Vashishtha et al., 2019; Zhou et al., 2019). In this paper …
Content truncated to 500 words
Labeled Text
AbstractKnowledge bases often consist of facts which are harvested from a variety of sources, many of which are noisy and some of which conflict, resulting in a level of uncertainty for each triple. Knowledge bases are also often incomplete, prompting the use of embedding methods to generalize from known facts, however, existing embedding methods only model triple-level uncertainty, and reasoning results lack global consistency. To address these shortcomings, we propose BEUrRE, a novel uncertain knowledge graph embedding method with calibrated probabilistic semantics. BEUrRE models each entity as a box (i.e. axis-aligned hyperrectangle) and relations between two entities as affine transforms on the head and tail entity boxes. The geometry of the boxes allows for efficient calculation of intersections and volumes, endowing the model with calibrated probabilistic semantics and facilitating the incorporation of relational constraints. Extensive experiments on two benchmark datasets show that BEUrRE consistently outperforms baselines on confidence prediction and fact ranking due to its probabilistic calibration and ability to capture high-order dependencies among facts.
Publication Details
- Paper ID:
- 21802
- Year:
- 2021
- Publication:
- TIPSTER
- Publication Type:
- conf
- DOI:
- nan