Do social media campaigns foster vaccination adherence? A systematic review of prior intervention-based campaigns on social media

The COVID-19 pandemic has demonstrated the importance of large-scale campaigns to facilitate vaccination adherence. Social media presents unique opportunities to reach broader audiences and reduces the costs of conducting national or global campaigns aimed at achieving herd immunity. Nonetheless, few studies have reviewed the effectiveness of prior social media campaigns for vaccination adherence, and several prior studies have shown that social media campaigns do not increase uptake rates. Hence, our objective is to conduct a systematic review to examine the effectiveness of social media campaigns and to identify the reasons for the mixed results of prior studies. Our methodology began with a search of seven databases, which resulted in the identification of 92 interventions conducted over digital media. Out of these 92 studies, only 15 adopted social media campaigns for immunization. We analyzed these 15 studies, along with a coding scheme we developed based on reviews of both health interventions and social media campaigns. Multiple coders, who were knowledgeable about social media campaigns and healthcare, analyzed the 15 cases and obtained an acceptable level of inter-coder reliability (> .80). The results from our systematic review show that only a few social media campaigns have succeeded in enhancing vaccination adherence. In addition, few campaigns have utilized known critical success factors of social media to induce vaccination adherence. Based on these findings, we discuss a set of research questions that informatics scholars should consider when identifying opportunities for using social media to resolve one of the most resilient challenges in public health. Finally, we conclude by discussing how the insights drawn from our systematic reviews contribute to advancing theories, such as social influence and the health belief model, into the realm of social media–based health interventions.     

Social network bursty topic discovery based on RNN and topic model

The data is noisy and diverse,with a large number of meaningless topics in social network.The traditional method of bursty topic discovery cannot solve the sparseness problem in social network,and require complicated post-processing.In order to tackle this problem,a bursty topic discovery method based on recurrent neural network and topic model was proposed.Firstly,the weight prior based on RNN and IDF were constructed to learn the relationship between words.At the same time,the word pairs were constructed to solve the sparseness problem.Secondly,the “spike and slab” prior was introduced to decouple the sparsity and smoothness of the bursty topic distribution.Finally,the burstiness of words were leveraged to model the bursty topic and the common topic,and automatically discover the bursty topics.To evaluate the effectiveness of proposed method,the various experiments were conducted.Both qualitative and quantitative evaluations demonstrate that the proposed RTM-SBTD method outperforms favorably against several state-of-the-art methods.     

NON-PARAMETRIC TOPIC MODEL FOR DISCOVERING GEOGRAPHICAL TOPIC VARIATIONS

This paper presents a non-parametric topic model that captures not only the latent topics in text collections, but also how the topics change over space. Unlike other recent work that relies on either Gaussian assumptions or diseretization of locations, here topics are associated with a distance dependent Chinese Restaurant Process （ddCRP）, and for each document, the observed words are influenced by the document＇s GPS-tag. Our model allows both unbound number and flexible distribution of the geographical variations of the topics＇ content. We develop a Gibbs sampler for the proposal, and compare it with existing models on a real data set basis.     

生活消费平台虚假评论识别模型的研究

生活消费平台已成为人们获取商家信息、反馈服务或产品质量的重要平台.虚假评论作为一种夸大或诽谤目标商家口碑的商业行为在生活消费平台很普遍,具有很强的危害性.本文对某网站的真实评论展开虚假评论研究,深入分析研究虚假评论的特征,从“可信度”的角度出发,提出用户及商家可信度模型.利用评论人的行为特征、商家的特征和评论文本的特征构建了虚假评论识别模型,经测试该模型达到了一个良好的识别效果.     

基于分层贝叶斯网络预测话题和单词复杂情感

In this paper, we provide a Word Emotion Topic (WET) model to predict the complex word emotion information from text, and discover the distribution of emotions among different topics. A complex emotion is defined as the combination of one or more singular emotions from following 8 basic emotion categories: joy, love, expectation, surprise, anxiety, sorrow, anger and hate. We use a hierarchical Bayesian network to model the emotions and topics in the text. Both the complex emotions and topics are drawn from raw texts, without considering any complicated language features. Our experiment shows promising results of word emotion prediction, which outperforms the traditional parsing methods such as the Hidden Markov Model and the Conditional Random Fields(CRFs) on raw text. We also explore the topic distribution by examining the emotion topic variation in an emotion topic diagram.     

基于主题相关性分析的文本倾向性研究

随着互联网的普及和电子商务的快速发展,网络评论、论坛讨论已成为人们网络生活的重要部分,并影响着社会舆论导向。如何识别网络评论对敏感主题（色情、法轮功等）的主观倾向性,把握网络舆情的正面或负面导向性,已成为信息安全领域研究的重要课题。文章以网络评论（影评）为研究对象,提出了一种分析文本语义倾向性的新模型,与传统倾向性识别系统不同的是,文章通过分析倾向性词汇与文本主题的相关性来研究文本的总体语义倾向。实验表明,新模型的判别准确率在80％以上,具有良好的应用前景。     

基于主题模型的（Aspect，Rating）摘要生成方法研究

提出基于短语参数学习的主题模型TMPP（Topic Model based on Phrase Parameter ）对在线评论中被评价实体的aspect和与之对应的rating进行抽取．TMPP具有三个特点：1）评论用“短语袋”表示；2）将标准的LDA中表示文档-主题的参数扩展为（aspect，rating）集；3）融合了先验知识．介绍了TMPP模型参数的物理含义、模型的生成过程以及先验知识的获取和表示方法；阐述了在TMPP模型中引入方面集聚类使用先验知识的原因与好处、TMPP模型提取（方面，等级）对形成（aspect，rating）摘要的原理．以真实的在线产品评论数据集为实验对象，在实验过程中引入先验知识的方面识别分析和等级预测精度分析，列出了五类产品相关方面和对立的情感词的实验结果．通过与已有的基线方法比较，实验表明若评论集中每篇评论有一个总体等级，TMPP能产生高质量的（aspect，rating）摘要．     

热丝CVD法沉积超薄α-Si∶H钝化膜

采用热丝化学气相沉积法在n型直拉单晶硅圆片表面双面沉积厚度为10 nm的本征非晶硅(α-Si∶H)薄膜.利用光谱型椭偏测试仪和准稳态光电导法研究热丝电流、H2体积流量和热丝与衬底之间的距离对α-Si∶H薄膜结构和钝化效果的影响.结果表明,热丝电流为21.5～23.5 A时,钝化后硅片的少子寿命随着热丝电流的增加呈现先增加后降低的趋势,热丝电流为23.0A时,钝化效果最好;H2体积流量为5～ 20 cm3/min时,少子寿命随着H2体积流量的增加呈现先增加后降低的规律,体积流量为15 cm3/min时,钝化效果最好;热丝与衬底间距为4～5 cm时,随着间距的增加,薄膜的结构由晶化向非晶化转变,在间距为4.5 cm时硅片的钝化效果达到最优.     

微博热点话题检测系统的设计与实现

本文描述了一个微博热点检测系统。管理者通过它可以快速了解正在发生的或是已发生的微博热点事件。系统采用调用微博API接口与改进爬虫程序相结合的方式获取网页数据;由于网络数据量巨大,为了提高效率,还采用了网页清理技术;重点介绍了话题活性模型的方法,系统可以根据时间坐标快速寻找热点话题,提高了热点话题发现的效率,大大降低了热点话题发现的时间复杂度。     

Special issue on aspects of spatial and temporal signal processing

Conclusions This special issue focuses on aspects of current research in a topic of paramount interest that spans a wide area of applications and at the same time offers theoretically challenging problems. I wish to thank Professor Sydney R. Parker for inviting me to serve as the Guest Editor for what he considered would be an important and timely special issue. I hope that expectation is, at least, partly realized by the readers of this volume. For me, it was a rewarding experience to have interacted with the contributors and reviewers. I also value this opportunity to study many interesting results generated over the past several years. The help provided by the reviewers are gratefully acknowledged. I wish to thank the Air Force Office of Scientific Research and the National Science Foundation for their support of research, some of which has been referenced in this editorial.     

BURST-LDA： A NEW TOPIC MODEL FOR DETECTING BURSTY TOPICS FROM STREAM TEXT

Topic models such as Latent Dirichlet Allocation （LDA） have been successfully applied to many text mining tasks for extracting topics embedded in corpora. However, existing topic models generally cannot discover bursty topics that experience a sudden increase during a period of time. In this paper, we propose a new topic model named Burst-LDA, which simultaneously discovers topics and reveals their burstiness through explicitly modeling each topic＇s burst states with a first order Markov chain and using the chain to generate the topic proportion of documents in a Logistic Normal fashion. A Gibbs sampling algorithm is developed for the posterior inference of the proposed model. Experimental results on a news data set show our model can efficiently discover bursty topics, outperforming the state-of-the-art method.     

Opt-out,abstain, unplug. A systematic review of the voluntary digital disconnection literature

Temporary and permanent disconnection from digital devices, platforms, or tools has gained traction from users and, subsequently, in academic discourse. A rapidly growing body of research focuses on so-called digital disconnection practices. However, the literature is highly scattered, with limited comprehensive work and consensus on essential foundations for this field. This study provides a systematic review of the digital disconnection literature following the PRISMA flow and Cochrane guidelines. We investigated 112 articles based on the following eight themes of digital disconnection: (1) definitions, (2) measurements, (3) prevalence, (4) motives, (5) strategies, (6) consequences/effectiveness, (7) relapsing, and (8) interventions. The review shows that research on this topic suffers from conceptual ambiguity and lacks consensus on terminology, definition, and measurement. As a first step to solving these lacunae, we provide a working definition, describing digital disconnection as a deliberate form of non-use of devices, platforms, features, interactions and/or messages that varies in frequency and duration with the aim of restoring or improving one’s perceived overuse, social interactions, psychological well-being, productivity, privacy and/or perceived usefulness. Moreover, we discuss the identified empirical and theoretical shortcomings and provide recommendations for future research.     

Single nucleotide polymorphism data analysis - State-of-the-art review on this emerging field from a signal processing viewpoint

The basic structural units of the genome are nucleotides. A single nucleotide polymorphism (SNP) is a mutation at a single nucleotide position. This paper discusses several major problems in SNP data analysis and review some existing solutions in this work. Generally speaking, a rich set of SNP analysis problems are cast in the signal processing framework. Our objective is to offer a state-of-the art review on this topic from a signal processing viewpoint so that researchers in the signal processing field can grasp the important domain knowledge to overcome the barrier between the two fields     

A Reliability-Program Case-History on Design Review

This paper summarizes the investigative results of actual design reviews as an important part of reliability program, and describes several reliability engineering efforts to achieve an effective design review. Design data packages (design documentation) which indicate the basic design program and design process are important in design reviews, When attention is concentrated on a data package, the ability of the reviewers is heightened and the results of the review are enhanced. When the design review is concerned with product reliability, then the availability and quality of: 1) a data package with established reliability level objectives and predictions, 2) a Failure Mode Effect Analysis and a Fault Tree Analysis, and 3) other data packages on product reliability and related technology or engineering, all greatly influence the results of the review. The potential weak points in a design can be revealed by over-stress tests and the results of such tests are very useful in the reliability design review. The improved design which can withstand the adequate over-stress tests appreciably lessened customer complaints about reliability.     

面向情感倾向性分类的特征表达研究

Online reviews and comments are important information resources for people. A new model, called Sentiment Vector Space Model (SVSM), for feature selection and weighting is proposed to predict the sentiment orientation of comments and reviews, e.g., sorting out positive reviews from negative ones. Different from that of topic oriented classification, feature selection of sentiment orientation prediction focuses on language characteristics. Different from traditional algorithms for sentiment classification, this model integrates grammatical knowledge and takes topic correlations into account. Features are extracted, and the similarity between these features and the topic are also computed. The feature similarity is taken as a factor when evaluating the polarity of opinions. The experimental results show that the proposed model is more effective in identifying sentiment orientation than most of the traditional techniques.     

Resolving the directness dilemma in document review sessions with nonnative speakers

Reviewers of technical documents must often work with nonnative speakers (NNSs) of English. Drawing on research in cross-cultural pragmatics and institutional discourse, we discuss linguistic patterns that document reviewers are likely to use when commenting on NNS writing. We anticipate miscommunications that may arise from some of these linguistic patterns, especially when a reviewer attempts to be both clear (so that the writer understands the comments) and polite (so that the reviewer maintains positive working relations with the writer). We recommend specific linguistic strategies that allow reviewers to balance clarity and politeness most effectively when communicating with NNSs.     

STOP, GO, and the state of the art in proposal writing

It is shown that the state of the art in proposal preparation makes available a wide array of techniques and devices to help make the proposal compliant, clear, convincing, and appealing. The techniques discussed are modular format, topical outlining, topic thesis sentences, required figures for topics, graphics oriented (GO) charts, figure enrichment, expanded figure titles, phrased topic titles, action topic titles, key issues visuals and lead topics, topic level storyboards, group wall review of storyboards, proposal manager's win strategy worksheets and customer's requirements worksheets, section level win strategy worksheets, compliance control system and worksheets, and early red team reviewing. In particular, the Sequential Topical Organization of Proposals (STOP) system, which introduced the modular format and topical storyboarding, is described     

基于干扰相似度的多话题演化模型

当前演化模型研究中,主要是单一话题在网络中的传播,较少考虑多话题之间的相互影响因素.在SIR模型的基础上提出了基于干扰相似度的多话题演化模型,该模型中的干扰是通过话题相似度对传播概率的影响来表征的.仿真结果表明,在临界值以内,正负两种趋向的话题相似度分别对话题演化的进程起到加强或阻碍作用,作用程度随着被干扰节点的度而变化,分别表现为正向相似度下的演化一致性和负向相似度下的演化分离性.超过临界值时,加强或阻碍作用均趋于饱和.     

社区环境下基于节点交互和主题的影响力计算模型

为解决现有算法对社交网络节点影响力计算准确度不高的问题，本文整合节点不同维度信息，综合考虑节点在多个主题社区上的主题分布向量，提出一种新的节点影响力计算模型.模型首先将主题相关性作为先验信息；然后利用混合隶属度随机块（Mixed Membership Stochastic Block）模型表达节点间的交互关系，用主题模型学习主题内容；最后结合全局拓扑关系迭代计算节点的全局影响力.本文选取社交网络数据，以P@N、MAP等作为评价指标同现有主流算法进行比较.实验结果显示，本文算法有效提升了影响力节点识别的准确度和排名的有效性.