基于SINA城市新闻的事件驱动情感原因提取数据集与方法

需积分: 10 65 浏览量更新于2024-08-12 收藏 435KB PDF 举报

本文主要探讨了"基于语料库构建的事件驱动情感原因提取"这一研究主题，发表于2016年的第十三届会议，自然语言处理实验方法(Natural Language Processing, NLP)上，地点位于美国德克萨斯州奥斯汀，时间是11月1日至5日。该研究由来自哈尔滨工业大学深圳研究生院、广东省级数据科学工程技术研究中心和香港理工大学计算机科学系的研究团队共同完成，作者包括Lin Gui、Dongyin Wu、Ruifeng Xu（共同第一作者）、Qin Lu和Yu Zhou。由于在情感原因提取领域缺乏公开可用的数据集，这限制了研究的发展。为解决这个问题，作者首先开发了一个新的数据集，该数据集来源于SINAcity新闻。他们使用的标注依据是W3C情绪标记语言(W3C Emotion Markup Language, EML)的框架，这是一种标准，用于描述和标记文本中的情绪表达和其可能的原因。研究的核心贡献在于提出了一种7元组定义，用于细致地描述情感引发事件。这个7元组模型可能包含了事件的主体、动作、时间、地点、情境、情感状态以及导致情感的原因等关键要素，有助于系统化和结构化地理解情感事件背后的原因。通过这种方式，研究者能够构建一个适用于机器学习和自然语言处理算法的情感原因提取模型，以自动从大量文本中识别和解析出情感触发的情境和因素。此外，论文还可能探讨了语料库建设的方法，如何选择和预处理数据，以及如何确保标注的准确性和一致性。可能还涉及了模型的训练策略、特征工程、以及评估指标的选择，如精确度、召回率或F1分数等，以衡量模型在情感原因提取任务上的性能。这篇研究论文不仅解决了情感原因提取领域的数据稀缺问题，还提出了一个创新的事件描述框架，并通过实际应用展示了在大规模文本数据中进行情感因果关系挖掘的可能性。这对于理解和分析公众情感动态、社交媒体舆情分析等领域具有重要的实践价值。

3 Construction of Emotion Cause Corpus

In this section, we ﬁrst describe the linguistic phe-

nomenon in emotion expressions. It serves as the in-

spiration to develop the annotated dataset. We then

introduce details of the annotation scheme, followed

by the construction of the dataset.

3.1 Linguistic Phenomenon of Emotion Causes

Emotion causes play an important role in emotion

expressions. An emotion cause reveals the stimulus

of an emotion. Considering linguistic phenomenon

of emotion causes, we follow three basic principles

in corpus construction: (1) Keep the whole context

of emotion expression; (2) The basic processing unit

is at the clause level; and (3) Use of formal text.

In written text, there is an emotion keyword,

which is used to express an emotion, in the context

of the emotion cause. Thus, ﬁnding the appropri-

ate context of emotion keywords in the annotation

is the pre-requisite to identify its cause. It is the

reason why we keep the whole context of emotion

keywords.

Another important kind of cues is the presence of

conjunctions and prepositions. These words indi-

cate the discourse information between clauses. In

order to make use of discourse information, the ba-

sic analysis unit should be at clause level rather than

at sentence level.

In the third principle, we choose the formal tex-

t in corpus construction. According to the related

works, emotion expressions can have overlapping e-

motion cause and emotion target (Gui et al., 2014)

in informal text. This is why some studies even in-

corporate cause extraction with target identiﬁcation

to improve performance. However, our focus is on

emotion cause identiﬁcation. We use formal news

text to avoid the potential mix up.

3.2 Collection and Annotation

We ﬁrst take 3 years (2013-15) Chinese city news

from NEWS SINA

containing 20,000 articles as the

raw corpus. Based on a list of 10,259 Chinese pri-

mary emotion keywords (keywords for short) (Xu

et al., 2008), we extract 15,687 instances by key-

word matching from the raw data. Here, we call the

presence of an emotion keyword as an instance in

http://news.sina.com.cn/society/

the corpus. For each matched keyword, we extract

three preceding clauses and three following clauses

as the context of an instance. If a sentence has more

than 3 clauses in each direction, the context will in-

clude the rest of the sentence to make the context

complete. For simplicity, we omit cross paragraph

context.

Note that the presence of keywords does not nec-

essarily convey emotional information due to differ-

ent possible reasons such as negative polarity and

sense ambiguity. For example, “祝愿/wishes” is

an emotion word of “happiness”. It can also be

the name of a song. Also, the presence of emotion

keywords does not necessarily guarantee the exis-

tence of emotional cause neither. After removing

those irrelevant instances, there are 2,105 instances

remain. For each emotional instance, two annotators

manually annotate the emotion categories and the

cause(es) in the W3C Emotion Markup Language

(EML) format. Ex1 shows an example of an anno-

tated emotional sentence in the corpus, presented by

the original simpliﬁed Chinese, followed by its En-

glish translation. To save space, we remove the xml

tags in the annotation. The original annotated data

is in a subsidiary ﬁle

. The basic analysis unit is a

clause. Emotion cause is marked by <cause>, and

the emotion keyword is marked by <keywords>. E-

motion type, POS, position and the length of anno-

tation are also annotated in Emotionml format.

Ex.1: 朱某今年55岁，1979年参加工作时

才19岁，已有36年的手艺。 “ 我当时被分到丹阳南

京理发店工作，这是当时丹阳最大的理发店。我在

那儿获得了好多证书和荣誉。 ”<cause POS=“v”

Dis=“-1”>说起自己的荣誉</cause>，朱某很

是<keywords type=happiness>自豪</keywords>。

Mr. Zhu is 55 years old. He started working

in 1979 as a barber when he was 19 , and has 36

years of experience. “I was assigned to work at the

Barbershop in Danyang, Nanjing. It is the largest

barbershop in Danyang. I won many awards and

honors there.” <cause POS=“v” Dis=“-1”>Talking

about his honors</cause>, Mr. Zhu is so <keywords

type=“happiness”> proud </keywords>.

Ex.1 only contains one cause. However, one key-

word may have more than one corresponding emo-

tion causes. In Ex.2, there are two relevant causes

http://hlt.hitsz.edu.cn/?page id=694

1641

剩余10页未读，继续阅读

weixin_38683562

粉丝: 6
资源: 970

基于SINA城市新闻的事件驱动情感原因提取数据集与方法

NLP文本分类语料库（复旦）语料训练集

tota：贸易协定语料库文本

大模型文本语料库之CnOpenDataA股上市公司公司章程公告数据样本数据

领域网站本体驱动的自动语料库构建与优化

Python驱动的旅游景点情感分析语料库与模型开发

自动获取翻译知识：基于词典的双语语料库对齐技术

机器翻译技术：双语语料库对齐与知识获取

国外隐喻语料库研究方法的深度剖析与展望

改进中英三元组可比语料库：理论与应用研究

机器翻译原理：双语语料库对齐与知识自动获取

最新资源