Who creates Trends in Online Social Media: The Crowd or Opinion Leaders?

Internet slang words can very quickly become ubiquitous because of social memes and viral online content. Weibo, a Twitter‐like service in China, demonstrates that the adoption of popular Internet slang undergoes 2 distinct peaks in its temporal evolution, in which the former is relatively much lower than the latter. An in‐depth comparison of the diffusion of these different peaks suggests that popular attention in the early stage of propagation results in large‐scale coverage, while the participation of opinion leaders at the early stage only leads to minor popularity. Our empirical results question the conventional influentials hypothesis and provide some insights for marketing practice and influence maximization in social networks.     

情感分析与认知

分析了情感分析的3个主要步骤,包括文本情感获取与表达、文本情感分类与计算以及文本情感分析的应用.情感分析得到的结论主要是对相关观点的摘要、对相关事件态度的预测或者统计等,但这些结论都没有发挥文本情感在认知中的作用.为了将情感分析应用于认知科学,提出了情感由情感信号和情感实体组成的观点.情感信号主要是指情感的一些形式载体,比如心跳加速、脸红等这些人体内外的某些表现,表达情感的文字、图片、声音等这类媒体.情感实体主要是指人类对情感形成的一种共识,比如爱、恨、憎恶、高兴、羞愧、嫉妒、内疚、恐惧、焦虑等与人的意识相关联的部分.同时提出了在人工智能中利用情感信息的设想.这对于模拟情感对认知的影响具有一定的意义.     

Studying users’ perception of IoT mobile companion apps

Internet of Things (IoT) products provide over-the-net capabilities such as remote activation, monitoring, and notifications. An associated mobile app is often provided for more convenient usage of these capabilities. The perceived quality of these companion apps can impact the success of the IoT product. We investigate the perceived quality and prominent issues of smart-home IoT mobile companion apps with the aim of deriving insights to: (i) provide guidance to end users interested in adopting IoT products; (ii) inform companion app developers and IoT producers about characteristics frequently criticized by users; (iii) highlight open research directions. We employ a mixed-methods approach, analyzing both quantitative and qualitative data. We assess the perceived quality of companion apps by quantitatively analyzing the star rating and the sentiment of 1,347,799 Android and 48,498 iOS user reviews. We identify the prominent issues that afflict companion apps by performing a qualitative manual analysis of 1,000 sampled reviews. Our analysis shows that users’ judgment has not improved over the years. A variety of functional and non-functional issues persist, such as difficulties in pairing with the device, software flakiness, poor user interfaces, and presence of issues of a socio-technical impact. Our study highlights several aspects of companion apps that require improvement in order to meet user expectations and identifies future directions.     

A Model for Cross-Domain Opinion Target Extraction in Sentiment Analysis

Opinion target extraction is one of the core tasks in sentiment analysis on text data. In recent years, dependency parser–based approaches have been commonly studied for opinion target extraction. However, dependency parsers are limited by language and grammatical constraints. Therefore, in this work, a sequential pattern-based rule mining model, which does not have such constraints, is proposed for cross-domain opinion target extraction from product reviews in unknown domains. Thus, knowing the domain of reviews while extracting opinion targets becomes no longer a requirement. The proposed model also reveals the difference between the concepts of opinion target and aspect, which are commonly confused in the literature. The model consists of two stages. In the first stage, the aspects of reviews are extracted from the target domain using the rules automatically generated from source domains. The aspects are also transferred from the source domains to a target domain. Moreover, aspect pruning is applied to further improve the performance of aspect extraction. In the second stage, the opinion target is extracted among the aspects extracted at the former stage using the rules automatically generated for opinion target extraction. The proposed model was evaluated on several benchmark datasets in different domains and compared against the literature. The experimental results revealed that the opinion targets of the reviews in unknown domains can be extracted with higher accuracy than those of the previous works.     

数字图书馆容灾系统建设研究

本文在阐述数据容灾概念与技术的基础上,简单分析了当前几种关键技术,并针对目前我国数字图书馆的可用性与安全性要求高的特点,提出了应用SAN技术实现数字图书馆容灾系统的设计方案,然后对此方案进行了分析,为其他容灾系统的建设起参考与借鉴作用。     

用户评价中比较句的识别和倾向性分析

在互联网上用户评价内容中很多比较句的比较结果反映了语句陈述者对比较对象的倾向性态度。根据已有的10类比较句句型总结了它们的常见概念搭配,在基于语义块的语句倾向性分析方法的基础上构建了比较句自动识别系统和比较句倾向性自动分析系统。采用第四届中文倾向性分析评测的语料进行实验,对语料中的比较句进行了识别,对比较句中的要素进行了抽取并且分析了比较句的倾向性,识别结果和倾向性分析结果均好于所有参评系统的平均值。     

穆青新闻传播思想刍议

通过对穆青新闻思想背景、新闻观念（媒介功能观、舆论观和新闻价值观）和新闻报道方式等方面进行阐述,透视穆青新闻传播思想特点：现实体验的文学化转换;政治语境下的媒体预设;超越事实的意义化表述,并以此挖掘其社会主义性质与新闻专业精神的内在矛盾,揭示其新闻观对后世新闻思想发展的引导价值及现实意义。     

一种基于情感倾向分析的网络团体意见领袖识别算法

意见领袖对网络舆情的产生和发展有着重要的指引作用,挖掘和识别网络社区中的意见领袖有重要的现实意义。结合聚类算法和分类算法的优势,提出一种基于话题内容分析的兴趣团体发现方法,以有效识别出兴趣团体。并通过分析用户回帖情感倾向来计算用户间链接的权重。在此基础上,提出了一种新的LeaderRank意见领袖发现算法,通过实验证明该算法能有效提高意见领袖挖掘的准确度。     

汉语语音合成系统评测中MOS和PC评估方法的比较

采用主观印象评测（MOS）和成对比较评测（PC）两种方法对已有的四个汉语语音合成系统进行了评测。17名普通话讲的很好的大学三年级学生参加了评测，其中男同学10名，女同学7名。在MOS中，文章从系统的整体印象、努力程度、理解程度、语速、发音（重音、停顿）及悦耳程度6个方面进行了评测。评测结果显示：听音的努力程度与系统的整体印象相关性最大，对系统整体印象的影响最大，其次是声音的悦耳程度。评测比较结果显示，PC测试同MOS测试具有很好的一致性，对于没有专家参与的评测中，PC方法更简单易用。     

高斯核模糊粗糙集中对象集变化时近似集增量更新方法研究

随着互联网技术的飞速发展,网络评论信息呈现爆炸式的增长,观点挖掘技术应运而生。评价对象与评价短语的抽取是观点挖掘中一项重要的任务。针对现有的基于模板方法存在人工参与过多、模板覆盖率不足、不能识别跨度较远的评价对象与评价短语等问题,提出了一种自动提取模板、利用概率进行评价搭配识别并能识别跨度较远的评价对象与评价短语的方法。通过引入同义词计算情感词的情感强度,综合考虑情感词与修饰词影响,完成倾向性判断。利用COAE2011的语料对上述方法进行了实验评价,并与两个baseline方法进行比较,取得了较好的实验结果。