Argument Inference from Relevant Event Mentions in Chinese Argument Extraction
Li, Peifeng and Zhu, Qiaoming and Zhou, Guodong

Article Structure

Abstract

As a paratactic language, sentence-level argument extraction in Chinese suffers much from the frequent occurrence of ellipsis with regard to inter-sentence arguments.

Introduction

The task of event extraction is to recognize event mentions of a predefined event type and their arguments (participants and attributes).

Related Work

Almost all the existing studies on argument extraction concern English.

Baseline

In the task of event extraction as defined in ACE evaluations, an event is defined as a specific occurrence involving participants (e.g., Person, Attacker, Agent, Defendant) and attributes (e.g., Place, Time).

Inferring Inter-Sentence Arguments on Relevant Event Mentions

In this paper, a global argument inference model is proposed to infer those inter-sentence arguments and their roles, incorporating with semantic relations between relevant event mention pairs and argument semantics.

Experimentation

In this section, we first describe the experimental settings and the baseline, and then evaluate our global argument inference model incorporating with relevant event mentions and argument semantics to infer arguments and their roles.

Conclusion

In this paper we propose a global argument inference model to extract those inter-sentence arguments due to the nature of Chinese that it is a discourse-driven pro-drop language with the wide spread of ellipsis and the open flexible sentence structure.

Topics

Coreference

Appears in 12 sentences as: corefered (1) Coreference (9) coreference (5)
In Argument Inference from Relevant Event Mentions in Chinese Argument Extraction
  1. To resolve such problem, this paper proposes a novel global argument inference model to explore specific relationships, such as Coreference , Sequence and Parallel, among relevant event mentions to recover those inter-sentence arguments in the sentence, discourse and document layers which represent the cohesion of an event or a topic.
    Page 1, “Abstract”
  2. extractor, it is really challenging to recognize these entities as the arguments of its corefered mention E3 since to reduce redundancy in a Chinese discourse, the later Chinese sentences omit many of these entities already mentioned in previous sentences.
    Page 2, “Introduction”
  3. In this paper, we divide the relations among relevant event mentions into three categories: Coreference , Sequence and Parallel.
    Page 4, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  4. An event may have more than one mention in a document and coreference event mentions refer to the same event, as same as the definition in the ACE evaluations.
    Page 4, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  5. Those coreference event mentions always have the same arguments and roles.
    Page 4, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  6. Therefore, employing this relation can infer the arguments of an event mention from their Coreference ones.
    Page 4, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  7. For example, we can recover the Time, Place and Instrument for E3 via its Coreference mention E2 in discourse D1, mentioned in Section 1.
    Page 4, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  8. The Coreference relation is divided into two types: Noun-based Coreference (NC) and Event-based Coreference (EC) while the former always uses a verbal noun to refer to an event mentioned in current or previous sentence and the latter is that an event is mentioned twice or more actually.
    Page 5, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  9. Coreference relation: Since the NC and EC relcation between two event mentions are different in the event expression, we introduce the discourse-based optimization for the former and document-based optimization for the latter.
    Page 6, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  10. To evaluate the similarity between two candidates a and a ’, we regard them as similar ones when they are the same word or in the same entity coreference chain.
    Page 7, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  11. We use a coreference resolution tool to construct the entity coreference chains, as described in Kong et al (2010).
    Page 7, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”

See all papers in Proc. ACL 2013 that mention Coreference.

See all papers in Proc. ACL that mention Coreference.

Back to top.

Chinese word

Appears in 6 sentences as: Chinese word (5) Chinese words (1)
In Argument Inference from Relevant Event Mentions in Chinese Argument Extraction
  1. The second issue is that the Chinese word order in a sentence is rather agile for the open
    Page 3, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  2. (2012a) find out that sometimes two trigger mentions are within a Chinese word whose morphological structure is Coordination.
    Page 4, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  3. The relation between those event mentions whose triggers merge a Chinese word or share the subject and the object are Parallel.
    Page 4, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  4. These constraints are enlightened by the fact that only Chinese words with Coordination structure can be divided into two new words and each word can trigger an event with the different event type 2 .
    Page 5, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  5. 1 It acts as the governing semantic element in a Chinese word .
    Page 5, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  6. Finally, all the sentences in the corpus are divided into words using a Chinese word segmentation tool (ICTCLAS)1 with all entities annotated in the corpus kept.
    Page 8, “Experimentation”

See all papers in Proc. ACL 2013 that mention Chinese word.

See all papers in Proc. ACL that mention Chinese word.

Back to top.

extraction system

Appears in 6 sentences as: extraction system (5) extraction systems (1)
In Argument Inference from Relevant Event Mentions in Chinese Argument Extraction
  1. Section 3 describes a state-of-the-art Chinese argument extraction system as the baseline.
    Page 2, “Introduction”
  2. As the baseline, we choose a state-of-the-art Chinese event extraction system , as described in Li et al.
    Page 3, “Baseline”
  3. For fair comparison, we adopt the same experimental settings as the state-of-the-art event extraction system (Li et al.
    Page 7, “Experimentation”
  4. Besides, all the experiments on argument extraction are done on the output of the trigger extraction system as described in Li et al.
    Page 8, “Experimentation”
  5. Table 3 shows the performance of the baseline trigger extraction system and Line 1 in Table 4 illustrates the results of argument identification and role determination based on this system.
    Page 8, “Experimentation”
  6. However, our model can be an effective complement of the sentence-level English argument extraction systems since the performance of argument extraction is still low in English and using discourse-level information is a way to improve its performance, especially for those event mentions whose arguments spread in complex sentences.
    Page 9, “Experimentation”

See all papers in Proc. ACL 2013 that mention extraction system.

See all papers in Proc. ACL that mention extraction system.

Back to top.

semantic relations

Appears in 5 sentences as: semantic relation (1) semantic relations (4)
In Argument Inference from Relevant Event Mentions in Chinese Argument Extraction
  1. 1) We propose a novel global argument inference model, in which various kinds of event relations are involved to infer more arguments on their semantic relations .
    Page 2, “Introduction”
  2. In this paper, a global argument inference model is proposed to infer those inter-sentence arguments and their roles, incorporating with semantic relations between relevant event mention pairs and argument semantics.
    Page 3, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  3. Therefore, employing those high level information to capture the semantic relation , not only the syntactic structure, between the trigger and its long distance arguments is the key to improve the performance of the Chinese argument identification.
    Page 4, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  4. Hence, the semantic relations among event mentions are helpful to be a bridge to identify those inter-sentence arguments.
    Page 4, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  5. We also introduce the argument semantics, which represent the semantic relations of argument-argument pair, argument-role pair and argument-trigger pair, to reflect the cohesion inside an event.
    Page 7, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”

See all papers in Proc. ACL 2013 that mention semantic relations.

See all papers in Proc. ACL that mention semantic relations.

Back to top.

sentence-level

Appears in 4 sentences as: sentence-level (4)
In Argument Inference from Relevant Event Mentions in Chinese Argument Extraction
  1. As a paratactic language, sentence-level argument extraction in Chinese suffers much from the frequent occurrence of ellipsis with regard to inter-sentence arguments.
    Page 1, “Abstract”
  2. In additional, there are only very few of them focusing on Chinese argument extraction and almost all aim to feature engineering and are based on sentence-level information and recast this task as an SRL-style task.
    Page 2, “Related Work”
  3. Liao and Grishman (2010) mainly focus on employing the cross-event consistency information to improve sentence-level trigger extraction and they also propose an inference method to infer the arguments following role consistency in a document.
    Page 2, “Related Work”
  4. However, our model can be an effective complement of the sentence-level English argument extraction systems since the performance of argument extraction is still low in English and using discourse-level information is a way to improve its performance, especially for those event mentions whose arguments spread in complex sentences.
    Page 9, “Experimentation”

See all papers in Proc. ACL 2013 that mention sentence-level.

See all papers in Proc. ACL that mention sentence-level.

Back to top.

development set

Appears in 3 sentences as: development set (3)
In Argument Inference from Relevant Event Mentions in Chinese Argument Extraction
  1. development set ; tri and tri ’ are triggers of kth and k’th event mention whose event types are et and et ’ in S<,~,j> and S<,~,,~> respectively.
    Page 7, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  2. Besides, we reserve 33 documents in the training set as the development set and use the ground truth entities, times and values for our training and testing.
    Page 8, “Experimentation”
  3. Our statistics on the development set shows almost 65% of the event mentions are involved in those Correfrence, Parallel and Sequence relations, which occupy 63%, 50%, 9% respectively6.
    Page 8, “Experimentation”

See all papers in Proc. ACL 2013 that mention development set.

See all papers in Proc. ACL that mention development set.

Back to top.

entity mention

Appears in 3 sentences as: entity mention (2) entity mentions (1)
In Argument Inference from Relevant Event Mentions in Chinese Argument Extraction
  1. The arguments are the entity mentions involved in an event mention with a specific role, the relation of an argument to an event where it participates.
    Page 3, “Baseline”
  2. is the kth event mentions in sentence S<U>; A<iyjykyl> is the lth candidate arguments in event mention T<U,k>; Z is used to denote <i,j,k,l>;f[(EZ) is the score of AI identifying entity mention EZ as an argument, where EZ is the lth entity of the kth event mention of the jth sentence of the ith discourse in document D. fD(EZ, Rm) is the score of RD assigning role Rm to argument E Z.
    Page 6, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”
  3. same entity mention .
    Page 6, “Inferring Inter-Sentence Arguments on Relevant Event Mentions”

See all papers in Proc. ACL 2013 that mention entity mention.

See all papers in Proc. ACL that mention entity mention.

Back to top.

ILP

Appears in 3 sentences as: ILP (3)
In Argument Inference from Relevant Event Mentions in Chinese Argument Extraction
  1. To achieve an optimal solution, we formulate the global inference problem as an Integer Linear Program ( ILP ), which leads to maximize the objective function.
    Page 8, “Experimentation”
  2. ILP is a mathematical method for constraint-based inference to find the optimal values for a set of variables that maximize an objective function in satisfying a certain number of constraints.
    Page 8, “Experimentation”
  3. In the literature, ILP has been widely used in many NLP applications (e.g., Barzilay and Lapata, 2006; Do et al., 2012; Li et al., 2012b).
    Page 8, “Experimentation”

See all papers in Proc. ACL 2013 that mention ILP.

See all papers in Proc. ACL that mention ILP.

Back to top.

PDF available from www.aclweb.org.
Send us an email