Exploring the Interactions of Storylines from Informative News Events

Today＇s news readers can be easily overwhelmed by the numerous news articles online. To cope with information overload, online news media publishes timelines for continuously developing news topics. However, the timeline summary does not show the relationship of storylines, and is not intuitive for readers to comprehend the development of a complex news topic. In this paper, we study a novel problem of exploring the interactions of storylines in a news topic. An interaction of two storylines is signified by informative news events that play a key role in both storylines. Storyline interactions can indicate key phases of a news topic, and reveal the latent connections among various aspects of the story. We address the coherence between news articles which is not considered in traditional similarity-based methods, and discover salient storyline interactions to form a clear, global picture of the news topic. User preference can be naturally integrated into our method to generate query-specific results. Comprehensive experiments on ten news topics show the effectiveness of our method over alternative approaches.     

基于图卷积网络的微博新闻故事线抽取方法

微博作为人们获取和传播新闻事件的主要平台,隐藏着丰富的事件信息。从微博数据中抽取故事线能为用户提供一种直观的方式来准确理解事件演化,然而微博数据稀疏和上下文缺乏的特点为故事线抽取带来了挑战。因此,通过两个连续的任务从微博数据中自动抽取故事线：1）基于微博传播影响力对事件进行建模,并提取出首要事件;2）基于事件特征建立异构事件图,提出事件图卷积网络（E-GCN）模型来提升对事件间隐式关系的学习能力,从而实现事件的故事分支预测并链接事件。在真实数据集上从故事分支和故事线两个角度进行评测,结果表明所提方法在故事分支生成测评中,相较于贝叶斯模型、斯坦纳树和故事森林在F1值上,在Dataset1上分别高出28个百分点、20个百分点和27个百分点,在Dataset2上分别高出19个百分点、12个百分点和22个百分点;而在故事线抽取评测中,相较于故事时间线、斯坦纳树和故事森林在正确的边准确率上,在Dataset1上分别高出33个百分点、23个百分点和17个百分点,在Dataset2上分别高出12个百分点、3个百分点和9个百分点。     

基于命名实体敏感的分层新闻故事线生成方法

社会网络中海量、无序且碎片化的新闻数据,使得人们无法从细粒度感知新闻事件,更无法多视角把握事件发展脉络。为了解决这个问题,该文提出基于命名实体敏感的分层新闻故事线生成方法,在无监督的情况下,充分利用新闻信息构造层次化、多视点的事件脉络。该方法主要通过以下3个步骤实现:(1)基于事件主题信息与隐式语义信息相结合的方法检测事件;(2)基于多维语义信息的社区检测算法划分主题事件的子事件;(3)基于多视点信息构造事件发展的脉络。在真实数据集上的实验结果表明,该方法在三个步骤比基线方法均有提高,其中在构造事件发展脉络阶段,该方法在理解性、概括性和准确性指标上分别高出0.44、0.11和0.50。     

基于全局/局部共现词对分布的汉越双语新闻事件线索分析

针对汉越双语新闻事件线索分析,提出了基于全局/局部共现词对分布的汉越双语事件线索生成方法。该方法首先将新闻话题词语分布作为全局词语表征全局事件,然后用一定时间粒度下新闻片段特有的时间、人物、地点等事件元素作为局部词语,分析新闻片段中全局词语和局部词语的共现关系,将全局/局部词语的共现规律作为监督信息,结合RCRP算法和汉越双语新闻的对齐语料,构建有监督话题生成主题模型,获得相应时间跨度下代表事件发展进程的子话题分布,通过子话题的分布反映事件发展的线索,从而构建出在线汉越双语事件线索生成模型。实验在汉越混合新闻数据集上进行,事件线索生成对比实验结果证明了提出的方法的有效性。
     

面向Web的故事脉络挖掘研究综述

互联网时代,纷繁复杂的Web信息使得人们难以快速、准确地获得新闻事件的故事脉络。因此,如何从Web信息中自动挖掘社会事件的故事脉络(简称“故事脉络挖掘”)成为近年来Web数据挖掘领域的一个研究热点。故事脉络挖掘旨在通过分析新闻事件与后续关联事件间的相互关系,抽取事件的演化阶段,并进一步挖掘事件的演化模式。故事脉络挖掘可应用于网络新闻检索、文本摘要、舆情监测等众多应用场景,具有重要的研究价值。该文首先概述了故事脉络挖掘的定义、流程及主要任务,然后从故事脉络构建和事件演化分析两个方面详细介绍了目前故事脉络挖掘方向的主要进展,接着比较了两类数据集及其评测标准,最后给出了故事脉络挖掘领域未来的若干研究挑战和技术框架。     

Effects of simulation‐based learning on students' statistical factual,conceptual and application knowledge

The purpose of this study was to (1) examine the effects of a storyline on learners' factual, conceptual and application knowledge with the use of a simulation for teaching introductory statistical skills and to (2) explore students' subjective enjoyment of various learning activities often used in statistics education. In order to conduct the study, two versions of a simulation were developed that differed in the presence or absence of a storyline attribute. Sixty‐four graduate students were randomly assigned to one of the two intervention conditions. Both intervention groups demonstrated significantly higher learning gains after interacting with the simulation. Particularly, both simulation‐based interventions had a positive significant effect on the acquisition of application knowledge and skills. However, no significant differences between the intervention groups on any learning outcome explored in the study were found. Results also showed that students rated the simulation used in the study as a more enjoyable learning activity in comparison to reading a textbook, lecture or teamwork. Students from the simulation without a storyline intervention reported higher enjoyment than the other intervention group. Implications of the findings for understanding the instructional benefits and shortcomings of embedding a storyline in digital learning content are discussed.     

中文微博故事线生成方法

新浪微博、腾讯微博等微博平台已经成为国内重要的网络媒体。随着海量的实时信息在微博上分享和传播,为每个用户提供更多方便,展现一目了然的实事资讯的任务已经迫在眉睫。这就需要在微博中理出重大事件的发展进程。该文中,我们将利用最小权重支配集和有向斯坦纳树在给定查询的微博数据集上生成故事线。该文的工作由三部分组成:第一部分是在Lucene检索出来的结果集上构建多视点图;其次,通过在图中寻找最小权重支配集来选出具有代表性的微博;最后,通过求解有向斯坦纳树问题来平滑地连接这些已挑选的微博,形成故事线。在实际数据集上的实验验证了该文提出系统的高效性和有效性。