共查询到20条相似文献，搜索用时 15 毫秒
1.
《Information processing & management》2020,57(3):102087
With the rapid development in mobile computing and Web technologies, online hate speech has been increasingly spread in social network platforms since it's easy to post any opinions. Previous studies confirm that exposure to online hate speech has serious offline consequences to historically deprived communities. Thus, research on automated hate speech detection has attracted much attention. However, the role of social networks in identifying hate-related vulnerable community is not well investigated. Hate speech can affect all population groups, but some are more vulnerable to its impact than others. For example, for ethnic groups whose languages have few computational resources, it is a challenge to automatically collect and process online texts, not to mention automatic hate speech detection on social media. In this paper, we propose a hate speech detection approach to identify hatred against vulnerable minority groups on social media. Firstly, in Spark distributed processing framework, posts are automatically collected and pre-processed, and features are extracted using word n-grams and word embedding techniques such as Word2Vec. Secondly, deep learning algorithms for classification such as Gated Recurrent Unit (GRU), a variety of Recurrent Neural Networks (RNNs), are used for hate speech detection. Finally, hate words are clustered with methods such as Word2Vec to predict the potential target ethnic group for hatred. In our experiments, we use Amharic language in Ethiopia as an example. Since there was no publicly available dataset for Amharic texts, we crawled Facebook pages to prepare the corpus. Since data annotation could be biased by culture, we recruit annotators from different cultural backgrounds and achieved better inter-annotator agreement. In our experimental results, feature extraction using word embedding techniques such as Word2Vec performs better in both classical and deep learning-based classification algorithms for hate speech detection, among which GRU achieves the best result. Our proposed approach can successfully identify the Tigre ethnic group as the highly vulnerable community in terms of hatred compared with Amhara and Oromo. As a result, hatred vulnerable group identification is vital to protect them by applying automatic hate speech detection model to remove contents that aggravate psychological harm and physical conflicts. This can also encourage the way towards the development of policies, strategies, and tools to empower and protect vulnerable communities. 相似文献
2.
News source evaluations based on fact-checking can help curb the consumption and spread of fake news on social media. Prior research has primarily considered source evaluations with intuitive icons that indicate whether or not news sources are reputable. But can we increase the power of these icons by adding more detailed information about the evaluation that explains the reasons for the icon? What additional benefit would such evaluation details bring? Would they have the same effect for both positive and negative evaluations? We conducted two online experiments to understand the effects of a source evaluation icon (a positive or negative summary of the evaluation) and more detailed evaluation information explaining the reasons for the icon. Our results show an asymmetric effect of positive and negative icons and details. Negative icons reduced the believability of the articles, but adding evaluation details supporting the icon had no additional effect. In contrast, positive icons had no significant effects, but adding evaluation details significantly increased believability. We also found that users were more likely to view the evaluation details when the content of the article aligned with their pre-existing opinions, but the valence of the icon (positive or negative) did not affect this decision. 相似文献
3.
【目的/意义】在线健康社区用户规模庞大,信息量浩如烟海,如何帮助社区管理者和用户判别有用信息,提
高决策效率是亟待解决的问题。【方法/过程】在复杂网络视角下,提出一个新的评论有用性分析框架。首先,采集
在线健康社区患者评论数据,采用文本分析法分析有用评论、非有用评论以及所有评论的主题分布和情感分布,初
步分析各类评论文本的有用性特征;其次,将各类评论文本分别转换为文本关联网络,使用社会网络分析方法进一
步分析其有用性特征;最后,分析评论有用性及其特征与患者发表评论、用户对评论的有用性投票以及文本关联网
络结构特征的关联性,实现基于文本关联网络的评论有用性分析。【结果/结论】有用评论和非有用评论文本关联网
络结构具有一定差异,在线健康社区用户就诊前后的信息需求和经验输出的重点有所不同。【创新/局限】基于复杂
网络视角研究在线健康社区评论有用性,但仅使用了好大夫在线的数据,未来可对更多数量和种类的在线健康社
区信息内容有用性进行研究。 相似文献
4.
《International Journal of Information Management》2016,36(3):425-432
Social commerce has evolved quickly in practice and gained attention in the IS discipline. However, trust has remained a vital component and is dominantly worth investigating. The purpose of this study, therefore, is to examine the roles of social commerce constructs and social support constructs (i.e., emotional support and informational support) in establishing trust on online community platforms. The study will apply the theoretical foundation of social commerce constructs proposed by Hajli. In order to provide a detailed understanding of the proposed model, a quantitative study involving a survey data gathered from online communities in Malaysia, including Facebook, Trip Advisor and LinkedIn was conducted. The data was analyzed and hypotheses were tested with structural equation modeling (SEM). Our results shed some lights on social commerce literature. The findings show that there are significant effect of social commerce constructs on social support, namely the emotional and informational support, and in turn, on trust- building. 相似文献
5.
《Information processing & management》2016,52(3):430-445
The explosion of online user-generated content (UGC) and the development of big data analysis provide a new opportunity and challenge to understand and respond to public opinions in the G2C e-government context. To better understand semantic searching of public comments on an online platform for citizens’ opinions about urban affairs issues, this paper proposed an approach based on the latent Dirichlet allocation (LDA), a probabilistic topic modeling method, and designed a practical system to provide users—municipal administrators of B-city—with satisfying searching results and the longitudinal changing curves of related topics. The system is developed to respond to actual demand from B-city's local government, and the user evaluation experiment results show that a system based on the LDA method could provide information that is more helpful to relevant staff members. Municipal administrators could better understand citizens’ online comments based on the proposed semantic search approach and could improve their decision-making process by considering public opinions. 相似文献
6.
《Information processing & management》2020,57(2):102031
Social media platforms allow users to express their opinions towards various topics online. Oftentimes, users’ opinions are not static, but might be changed over time due to the influences from their neighbors in social networks or updated based on arguments encountered that undermine their beliefs. In this paper, we propose to use a Recurrent Neural Network (RNN) to model each user’s posting behaviors on Twitter and incorporate their neighbors’ topic-associated context as attention signals using an attention mechanism for user-level stance prediction. Moreover, our proposed model operates in an online setting in that its parameters are continuously updated with the Twitter stream data and can be used to predict user’s topic-dependent stance. Detailed evaluation on two Twitter datasets, related to Brexit and US General Election, justifies the superior performance of our neural opinion dynamics model over both static and dynamic alternatives for user-level stance prediction. 相似文献
7.
On exploring the impact of users’ bullish-bearish tendencies in online community on the stock market
《Information processing & management》2020,57(5):102209
The online financial community enables non-professional individual investors to express opinions, share information and even spread emotions through the Internet. This paper uses 5,178,824 comments published in an online financial community to study the users’ bullish-bearish tendencies on the stock market. To that end, we propose a convolutional neural network based classifier to extract users’ tendencies from their comments, and introduce the distributed lag model and the GARCH model to investigate the impact of users’ tendencies on market volatility and market returns. The results show that the online users’ bearish tendencies are reflected in stronger market volatility and higher market returns, and the consistency of online users’ tendencies has a positive impact on market volatility. 相似文献
8.
Jörgen Skågeby 《The Information Society》2013,29(1):60-72
Social metadata are receiving interest from many domains, mainly as a way to aggregate various patterns in social networks. Few scholars have, however, taken the perspective of end users and examined how they utilize social metadata to enrich interpersonal communication. The results of a study of end-user practices of social metadata usage are presented in this article. Data were gathered from a variety of online forums by collecting and analyzing user discussions relating to social metadata supporting features in Facebook. Three hundred and fifteen relevant comments on social metadata usage were extracted. The analysis revealed the use of experimental profiles, clashes between work-and non-work-related social metadata usage and differences in users' social investment, causing social dilemmas. The study also resulted in developments of theory relating to social metadata and relationship maintenance. In conclusion, social metadata expand a pure “attention economy,” conveying a much wider qualitative range of social information. 相似文献
9.
同侪影响对用户贡献行为的作用研究———基于网络客观大数据的分析 《科学学研究》2021,39(12):2294-2304
本文将同侪影响引入在线创新社区的用户行为研究中,从广度和深度两方面考察同侪影响对用户贡献行为的影响,并分析感知收益的中介作用。研究以小米社区MIUI功能与讨论区的创意集市板块为对象构建S-O-R模型,采用6567名用户发布的8830条创意、5.26万条评论和收到的103.36万条评论数据,利用Mplus8.1分析检验,结果发现：同侪影响广度与深度均有利于促进用户贡献行为,综合收益在同侪影响广度、深度与用户贡献行为间起正向中介效应,情感收益仅在同侪影响广度、深度与主动贡献行为间起正向中介效应,而认知收益则在同侪影响深度与反应贡献行为间起负向中介效应。研究拓展了在线网络情境下知识管理与社会学领域的交叉研究,并为在线创新社区社交网络和知识管理提供重要启示。 相似文献
10.
Veera Kangaspunta 《The Information Society》2013,29(5):275-288
AbstractThe research on online news comments has been dominated by a normative approach and has centered on media engagement. Normativity and media dominance have also featured big in the theoretical discussions on the public sphere. This article presents a case study of online news comments, combining a novel methodological testing of social network hypotheses to examine user–user interactions in online comments with a conceptual discussion of the potential connections between social network research and theories of the public. The social network analysis in this study indicated that users (online commentators) do not constitute highly dense networks, although their relations can be studied as social networks. However, this analysis can only explore limited features of this online phenomenon and requires complementary methods. From a conceptual perspective, this article confirms the role of shared issue for a potential public and also emphasizes the importance of context, actors, and meanings for understanding the public. 相似文献
11.
The present paper explores the relationship between nature of Facebook usage, non-directional comparisons and depressive syndromes. The extant research on linkage between social media usage and mental health is inconclusive. The paper uses data collected through an online survey of 399 Facebook users in the UK. A Facebook frequency rating scale was developed and validated as a part of the study. The Iowa-Netherlands Comparison Orientation Measure was modified and used to measure social comparison. The depressive syndromes were captured by the modified Center for Epidemiological Studies Depression Scale. The Rank Theory of Depression was used a guiding framework. The data collection had focused on the 20–29 year olds, as justified by the literature.The study found a negative relationship between active Facebook use and non-directional social comparisons. The relationship was reversed in the case of passive usage. There is small but significant causal linkage between increased non-directional social comparisons and depressive symptoms among the users. 相似文献
12.
Engin Bozdag 《Ethics and Information Technology》2013,15(3):209-227
Online information intermediaries such as Facebook and Google are slowly replacing traditional media channels thereby partly becoming the gatekeepers of our society. To deal with the growing amount of information on the social web and the burden it brings on the average user, these gatekeepers recently started to introduce personalization features, algorithms that filter information per individual. In this paper we show that these online services that filter information are not merely algorithms. Humans not only affect the design of the algorithms, but they also can manually influence the filtering process even when the algorithm is operational. We further analyze filtering processes in detail, show how personalization connects to other filtering techniques, and show that both human and technical biases are present in today’s emergent gatekeepers. We use the existing literature on gatekeeping and search engine bias and provide a model of algorithmic gatekeeping. 相似文献
13.
14.
The findings of our experiments showed that social network sites (SNSs) such as Google Plus, Facebook, and Twitter, have the ability to acquire knowledge about their users’ movements not only within SNSs but also beyond SNS boundaries, particularly among websites that embedded SNS widgets such as Google’s Plus One button, Facebook’s Like button, and Twitter’s Tweet button. In this paper, we analysed the privacy implication of such a practice from a moral perspective by applying Helen Nissenbaum’s decision heuristic derived from her contextual integrity framework in order to answer the question of whether or not an online user’s privacy is being violated by this practice. 相似文献
15.
《Information processing & management》2019,56(6):102079
The emergence of social media and the huge amount of opinions that are posted everyday have influenced online reputation management. Reputation experts need to filter and control what is posted online and, more importantly, determine if an online post is going to have positive or negative implications towards the entity of interest. This task is challenging, considering that there are posts that have implications on an entity's reputation but do not express any sentiment. In this paper, we propose two approaches for propagating sentiment signals to estimate reputation polarity of tweets. The first approach is based on sentiment lexicons augmentation, whereas the second is based on direct propagation of sentiment signals to tweets that discuss the same topic. In addition, we present a polar fact filter that is able to differentiate between reputation-bearing and reputation-neutral tweets. Our experiments indicate that weakly supervised annotation of reputation polarity is feasible and that sentiment signals can be propagated to effectively estimate the reputation polarity of tweets. Finally, we show that learning PMI values from the training data is the most effective approach for reputation polarity analysis. 相似文献
16.
在网络社区兴起的背景下,鉴于网络社区的海量评论数据中蕴含着大量专家用户群体智慧,本文提出基于网络评论文本挖掘的技术预见新型方法,以促进技术预见活动顺利实施并取得准确可信的最终结果。首先从多源数据中获得种子科技主题,并将其投放至开放网络社区,吸引专家用户进行充分讨论形成交互数据,经过数据爬取、清洗、存储等环节得到网络评论数据集,再利用情感分析、主题模型等方法对网络评论中蕴含的隐性知识进行显性化挖掘,并结合相关领域专家的研判,最终得到辅助技术预见决策的有价值信息。通过新型方法,可以使技术预见活动大幅降低成本、打破时空限制,便于大规模专家参与其中,并最大限度降低少数专家主观色彩浓厚的负面影响。 相似文献
17.
《Information processing & management》2020,57(4):102236
Online peer-to-peer (P2P) lending has developed dramatically over the last decade in China. But this rapid boom carries potential risks. Investors have incurred incalculable losses due to the recent increase in fraudulent and/or unreliable online P2P platforms. Hence, predicting and identifying potential default risk platforms is crucial at this juncture. To achieve this end, we propose a two-step method which employs a deep learning neural network to extract keywords from investor comments and then utilizes a bidirectional long short-term memory (BiLSTM) based model to predict the default risk of platforms. Experimental results on real-world datasets of about 1000 platforms show that in the keyword extraction phase, our model can better capture semantic features from highly colloquial comment-text and achieve significant improvement over other baselines. Additionally, in the default platform prediction stage, our model achieves an F1 value of 80.34% in identifying potential problem platforms, outperforming four baselines by 23.37%, 5.71%, 8.93%, and 4.98% of improvement and comprehensively verifying the effectiveness of our method. Our study provides an alternative solution for platform default risk prediction issues and validates the effectiveness of investor comments in revealing the risk situation of online lending platforms. 相似文献
18.
Seyed Amin Mirlohi Falavarjani Fattane Zarrinkalam Jelena Jovanovic Ebrahim Bagheri Ali A. Ghorbani 《Information processing & management》2019,56(6):102070
The ever increasing presence of online social networks in users’ daily lives has led to the interplay between users’ online and offline activities. There have already been several works that have studied the impact of users’ online activities on their offline behavior, e.g., the impact of interaction with friends on an exercise social network on the number of daily steps. In this paper, we consider the inverse to what has already been studied and report on our extensive study that explores the potential causal effects of users’ offline activities on their online social behavior. The objective of our work is to understand whether the activities that users are involved with in their real daily life, which place them within or away from social situations, have any direct causal impact on their behavior in online social networks. Our work is motivated by the theory of normative social influence, which argues that individuals may show behaviors or express opinions that conform to those of the community for the sake of being accepted or from fear of rejection or isolation. We have collected data from two online social networks, namely Twitter and Foursquare, and systematically aligned user content on both social networks. On this basis, we have performed a natural experiment that took the form of an interrupted time series with a comparison group design to study whether users’ socially situated offline activities exhibited through their Foursquare check-ins impact their online behavior captured through the content they share on Twitter. Our main findings can be summarised as follows (1) a change in users’ offline behavior that affects the level of users’ exposure to social situations, e.g., starting to go to the gym or discontinuing frequenting bars, can have a causal impact on users’ online topical interests and sentiment; and (2) the causal relations between users’ socially situated offline activities and their online social behavior can be used to build effective predictive models of users’ online topical interests and sentiments. 相似文献
19.
《International Journal of Information Management》2016,36(3):414-424
Cyberbullying by way of malicious online comments has been identified as a critical social issue. As a way to combat cyberbullying, it is important to promote the posting of benevolent comments, especially in response to malicious comments. In this study, we adopted a mixed methods approach in using social exchange theory to explore what motivates people to post benevolent comments online. We first adopted a qualitative study to explore the decision factors involved. These were then used as the basis for development of a theoretical research model to undertake a quantitative study. The results explain how people decide to post benevolent comments online. The study makes a strong theoretical contribution in demonstrating the decision factors underlying the posting of benevolent comments. It also has practical implications by providing guidance on how to combat cyberbullying through promoting the posting of benevolent online comments. 相似文献
20.
We empirically explore the associations between social media use at home and shopping preferences using survey data. We focus on popular retail firms including brick-and-mortar firms such as Walmart, Target, Nordstrom, and Best Buy, and online retailers, such as Amazon, Walmart, Target, and Best Buy. Social media use of popular platforms such as Facebook, Twitter, LinkedIn, Skype and a general category Other Social Media are analyzed. We find that use of LinkedIn, Skype and Other Social Media at home, in the model without control variables, is associated with shopping at Nordstrom, Walmart and Target. Shopping online at Amazon, Best Buy and Walmart, without control variables in the model specification, is associated with use of Facebook, Skype, Twitter and Other Social Media at home. We report additional insights using an alternative specification that includes social media use at work. Media Richness Theory (MRT) and Strength of Weak Ties from Social Network Analysis (SNA) and related theories help explain our results. Our results have important implications for social marketing campaigns and social media policies for consumer retail firms. 相似文献