NgramPOS: a bigram-based linguistic and statistical feature process model for unstructured text classification

Research in financial domain has shown that sentiment aspects of stock news have a profound impact on volume trades, volatility, stock prices and firm earnings. In-depth analysis of stock news is now sourced from financial reviews by various social networking and marketing sites to help improve decision making. Nonetheless, such reviews are in the form of unstructured text, which requires natural language processing (NLP) in order to extract the sentiments. Accordingly, in this study we investigate the use of NLP tasks in effort to improve the performance of sentiment classification in evaluating the information content of financial news as an instrument in investment decision support system. At present, feature extraction approach is mainly based on the occurrence frequency of words. Therefore low-frequency linguistic features that could be critical in sentiment classification are typically ignored. In this research, we attempt to improve current sentiment analysis approaches for financial news classification by focusing on low-frequency but informative linguistic expressions. Our proposed combination of low and high-frequency linguistic expressions contributes a novel set of features for sentiment classification. The experimental results show that an optimal Ngram feature selection (combination of optimal unigram and bigram features) enhances sentiment classification accuracy as compared to other types of feature sets.

     

基于网络社交的开源情报研究与应用

网络社交已成为日常生活不可缺少的一部分,网络社交情报作为一种情报搜集的新形式,已经成为开源情报重要情报来源之一。探讨这一概念将有助于我们理解和把握互联网环境下的社交媒体作为国家情报源的重要价值。首先研究了网络社交中获取开源情报的途径,然后分析了网络社交的情报价值和例举了网络社交在开源情报中的应用,最后阐述了网络社交情报面临的挑战。结果发现,网络社交虽为一个极为诱人的情报源泉,但又是一个巨大的信息安全陷阱,各类网络社交平台也将成为未来情报战的又一个主战场。     

基于级联模型的中文情感要素抽取

随着社交媒体的发展及成熟,每天在互联网环境中都会产生大量的用户评论信息。抽取评价短语、评价对象和观点持有者等情感要素,已经成为了中文观点挖掘和情感分析的重要先决任务。针对中文情感要素抽取任务,本文提出了一个统计和规则相结合的级联模型,主要贡献包括：（1）针对汽车领域评论信息,构建情感要素标注语料库和相关词典;（2）对于以往研究较少关注的中文评价短语,本文详细分析阐述其定义和分类;（3）结合统计和规则,分别针对评价短语和情感要素提出级联抽取策略。实验结果充分证明了该级联模型的有效性,相比较于其它基于规则的情感要素抽取算法有效提升了召回率,同时为后续社交媒体情感分析任务提供了有力的支持。     

Optimal consumption decisions of family networks with similar felicity functions

Family networks have increased at an accelerated pace through social networking through social networking services, online platforms, and social media clubs. This paper develops a model of an economy populated by networks of families with similar preferences (the same functional form of the felicity or utility function) but family members differ in their deep parameters (compulsion for consumption and risk tolerance), which, in turn, leads to some kind o heterogeneity among the families. In this context, the problem of felicity (utility) maximization of the average family in the network is solved. The optimal path of consumption is obtained for the average family and graphical comparative static exercises are performed. Finally, the economic welfare of the typical family in a network is calculated and comparative static experiments are carried out. This approach improves the understanding of the consumption decision making of families interacting in social networking services.

     

面向顾客点评数据的属性层次观点挖掘研究

This paper focuses on how to improve aspect-level opinion mining for online customer reviews. We first propose a novel generative topic model, the Joint Aspect/Sen-timent (JAS) model, to jointly extract aspects and aspect-dependent sentiment lexicons from online customer reviews. An as-pect-dependent sentiment lexicon refers to the aspect-specific opinion words along with their aspect-aware sentiment polarities with respect to a specific aspect. We then apply the extracted aspect- dependent sentiment lexi-cons to a series of aspect-level opinion mining tasks, including implicit aspect identification, aspect-based extractive opinion summarization, and aspect-level sentiment classification. Experimental results demonstrate the effectiveness of the JAS model in learning aspect- dependent sentiment lexicons and the practical values of the extracted lexicons when applied to these practical tasks.     

The collective storytelling organisational framework for social media use

The organisational implementation and use of social media for online community engagement is plagued with many challenges. The result is managers being frustrated at not achieving their desired results. Despite an increase in social media research, it is evident that some critical aspects of social media use by organisations are left unexplained by existing theories. The purpose of this research study is to investigate how social media is used within organisations to engage with its online community. The grounded theory method (GTM) was used to develop a substantive theory that increases understanding of this phenomenon. Two organisations were selected as the case studies. Both are industry leaders in South Africa, with one being a prominent retailer with a visible social media presence and the other, being a leading university in South Africa, which is actively growing its social media presence. The results show that organisations enter a collective storytelling process with their online community. The need to reduce risk to reputation and improve on the levels of online community engagement are identified as reasons for this. The strategic alignment of the social media function and the organisational social media use characteristics set the backdrop for the collective storytelling process, while content quality influences the richness of this process. Challenges associated with the collective storytelling process lead to the development and refinement of the organisational social media strategy, which in turn helps to reduce risk to reputation and improve online community engagement.     

The influence of online forum and SNS use on online political discussion in China: Assessing “Spirals of Trust”

Echoing the significance of mobile online networks in fueling the Arab Spring, the present study seeks to better understand social media influences in China by studying political activity among Chinese netizens. A survey of Chinese college students examines the influence of online social networks in the context of political attitudes and political participation. Study results reveal a moderate but positive impact of online forum and social networking site use on online political discussion. Implications for political change in the social networking era, particularly in regimes that practice Internet censorship like China’s, are discussed.     

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.     

The future of online social networks (OSN): A measurement analysis using social media tools and application

The explosion of online social networks (OSN) has created an interactive and communicative global phenomenon that has enabled billions of users to connect to other individuals on Facebook and Twitter but also with media sharing platforms such as Instagram and Pinterest. This study examines the current use of social media platforms and explores the factors that help define the long term implications of social media. The study employed a nationwide survey collected from 2012 to 2013 and is available from the PEW Internet research center of more than 2000 American citizens’ behaviour towards OSNs. The results revealed strong predictors of OSN that form the connections among users; and the core significant predictors: age, gender and access to mobile Internet that foster the adoption and usage of OSN in the future. Furthermore, online activities such as posting video content on social networks also highlighted the online usage patterns and trends of using social media to actively engage with other users more willingly than text. This is due to the viral nature of online media sharing on social media and as part of the video viewing and creating experience. An outline of practical implications of the findings and areas for future research is also discussed.     

Bubbles bursting: Investigating and measuring the personalisation of social media searches

Social media platforms implement personalisation algorithms to provide users with a tailored selection of posts under the assumption of a better experience. However, prior studies examining social media timelines revealed that, due to personalisation algorithms, social media users are more likely to encounter attitude-consistent content that reinforces their existing beliefs than information that contradicts them, creating filter bubbles and ultimately hampering their ability to make good decisions. To burst the bubbles, this paper proposes a framework for investigating and measuring the factors that affect personalisation in social media search mechanisms by controlling external noises that can mask the results and be misinterpreted as personalisation. Upon conducting a comprehensive set of experiments with Twitter as our study social media platform, we observed that users’ followees, cookies, and carry-over effect play a role in shaping personalised search results. While the extent of their influence is relatively limited, they can still lead to the introduction of biases, consequently affecting users’ opinions and judgements. In addition, the experiments also shed light on the fact that searches on polarised issues can lead to unwarranted one-sided inclinations in the results. Lastly, we argue in favour of alleviating the multifaceted societal and ethical consequences of algorithmic personalisation through encouraging users to develop healthy social media use habits and by urging social media platforms as well as policymakers to actively work towards fostering a more diverse and reliable online information ecosystem.     

Assessing the helpfulness of online hotel reviews: A classification-based approach

With the rapid development of Web 2.0, travelers have started sharing their travel experiences on websites. The expanding amount of online hotel reviews results in the problem of information overload. Therefore, the effective identification of helpful reviews has become an important research issue. In this study, online hotel reviews were collected from TripAdvisor.com, and the helpfulness of these reviews was comprehensively investigated from the aspects of review quality, review sentiment, and reviewer characteristics. Review helpfulness prediction models were also developed by using classification techniques. The results indicate that reviewer characteristics are good predictors of review helpfulness, whereas review quality and review sentiment are poor predictors of review helpfulness.     

Traces of mobility: Examining location disclosure on social networks with mobile location tagging

The prevailing adoption of mobile devices has facilitated individuals’ engagement with location tagging; thus, location-based information sharing, which allows geolocation specificity and interaction on the Internet platforms, has become a popular activity in the space of social media. This spatiotemporal information provides a great potential opportunity for marketers as photos or videos shared among friends on social networking sites can be viewed as a type of word-of-mouth that may increase brand awareness and attract more users. Thus, understanding the underlying factors driving location tagging intentions provides significant value for practitioners. However, despite its importance, our understanding of this topic is limited. This research draws from social exchange theory and the theory of impression management to exemplify the underlying factors driving individuals to engage in photo or video sharing with location tagging on social media. Using an online survey, we recruited 402 users. We find that location tagging intention is driven by a facilitating route involving social benefits and functional benefits and an impeding route involving perceived risk related to privacy concerns. Our findings provide important academic and managerial implications.     

Social by definition: How users define social platforms and why it matters

Social media platforms are characterized by diverse features and functions, and these facets remain in constant flux over time. This research examines how users define the central purpose of four major platforms in the United States (Facebook, Twitter, Instagram, Snapchat), and how such lay definitions relate to key outcomes previously associated with social media use. In Study 1, we validated self-report measures using a comparative scaling approach to capture what users view as the most defining categories of the four platforms. In Study 2, we investigated whether lay definitions of platforms relate to perceptions of social affordances and social resources. Overall, results provided evidence that defining platforms as social interaction (vs. other categories) is associated with amplified social affordances and resources. Together, the studies contribute to our understanding of how users navigate a dynamic online ecosystem, as well as how lay definitions may anchor the experiences and effects of social media.     

Identification of key cyberbullies: A text mining and social network analysis approach

Cyberbullying is a major problem in society, and the damage it causes is becoming increasingly significant. Previous studies on cyberbullying focused on detecting and classifying malicious comments. However, our study focuses on a substantive alternative to block malicious comments via identifying key offenders through the application of methods of text mining and social network analysis (SNA). Thus, we propose a practical method of identifying social network users who make high rates of insulting comments and analyzing their resultant influence on the community. We select the Korean online community of Daum Agora to validate our proposed method. We collect over 650,000 posts and comments via web crawling. By applying a text mining method, we calculate the Losada ratio, a ratio of positive-to-negative comments. We then propose a cyberbullying index and calculate it based on text mining. By applying the SNA method, we analyze relationships among users so as to ascertain the influence that the core users have on the community. We validate the proposed method of identifying key cyberbullies through a real-world application and evaluations. The proposed method has implications for managing online communities and reducing cyberbullying.     

Instagram versus Snapchat: Self-expression and privacy concern on social media

With the increasing popularity of image-based social networking applications (apps), such as Snapchat and Instagram, social media users express multifaceted aspects of their self-concept online, while remaining concerned about privacy issues. This study investigates the determinants behind the choice to use a social networking site (SNS) platform and predicts that the type of self an individual wants to express through the platform and their level of privacy concerns will play an important role in this choice. A total of 547 SNS users completed an online survey. Expression of multiple aspects of the self – true, actual, and ideal self – and online privacy concerns were empirically tested as the key determinants of SNS platform choice. The findings show that the expression of true and actual self are significant determinants for using Snapchat as a primary SNS, whereas the expression of ideal self and greater privacy concerns are associated with active Instagram use. The authors discuss the implications of these findings and recommend future research.     

Predictors of television and online video platform use: A coexistence model of old and new video platforms

Recognizing that new and old media coexist in media markets, the overarching aim of this study is to investigate how the perceived characteristics of online video platforms and consumer-related factors affect consumer intention to use the Internet and television to watch video content. The primary theoretical foundations are the theory of planned behavior (TPB) and the technology acceptance model (TAM). By extending TAM and TPB into other constructs, the present study aims to provide richer explanations for consumers’ choice of a video platform in the competitive video marketplace. This study used a survey method to collect data. A total of 1500 adults throughout the US who use the Internet were employed for the sample of the main survey. For the analysis to test hypotheses, 388 responses were used. This study found that the more consumers perceive online video platforms differ from television in satisfying their needs, the more likely they are to use online video platforms. The relative advantage and compatibility of online video platforms decrease the likelihood of using television.     

#Buzz: Exploring public interest in pollinators,bees, and honey using online media data

Pollinators play an important role in agriculture, food security, and ecosystems. They are facing serious population declines in the US as well as around the world, leading to concerns regarding natural fertilization, reduced crop yields, and economic threats to farmers and/or related agricultural businesses. Governments and non-governmental organizations have introduced policies and programs to conserve pollinators and increase public awareness of pollinator decline. This study focuses on online and social media content about pollinators as a proxy for public interest, measured by volume and sentiment of media. Additionally, we evaluate potential links between public interest and the reaction of markets of pollinator-derived products, namely honey. Volume and sentiment of online media surrounding pollinators was quantified for the years 2017 – 2021. Search results show mentions mostly driven by pleasant mentions of bees, honey, honeybee costumes, and butterfly migration rather than topics related to the pollination-related threats, causes, polices, and/or programs. Wholesale and retail prices of honey have been less variable than mentions over time, suggesting no apparent relationship between volume of online media activities surrounding pollinators (including bees) and honey prices. The low volume of mentions about threats towards pollinators may suggest a low level of public awareness of the topic and yield concerns for the effectiveness of polices for protecting pollinators and/or concern for economics of beekeepers, even with the active legislations of the US federal and states government for promoting pollinators health.     

Recommendation agents and information sharing through social media for coronavirus outbreak

The novel outbreak of coronavirus disease (COVID-19) was an unexpected event for tourism in the world as well as tourism in the Netherlands. In this situation, the travelers’ decision-making for tourism destinations was heavily affected by this global event. Social media usage has played an essential role in travelers’ decision-making and increased the awareness of travel-related risks from the COVID-19 outbreak. Online consumer media for the outbreak of COVID-19 has been a crucial source of information for travelers. In the current situation, tourists are using electronic word of mouth (eWOM) more and more for travel planning. Opinions provided by peer travelers for the outbreak of COVID-19 tend to reduce the possibility of poor decisions. Nevertheless, the increasing number of reviews per experience makes reading all feedback hard to make an informed decision. Accordingly, recommendation agents developed by machine learning techniques can be effective in the analysis of such social big data for the identification of useful patterns from the data, knowledge discovery, and real-time service recommendations. The current research aims to adopt a framework for the recommendation agents through topic modeling to uncover the most important dimensions of COVID-19 reviews in the Netherland forums in TripAdvisor. This study demonstrates how social networking websites and online reviews can be effective in unexpected events for travelers’ decision making. We conclude with the implications of our study for future research and practice.     

Drivers of social media fatigue: A systematic review

With a large amount of time spent on social media platforms, more and more people are suffering from social media fatigue. The concept of social media fatigue refers to a host of negative emotional responses to activities on social networking sites, such as tiredness, burnout, exhaustion, frustration, disinterest toward communication. Since research on social media fatigue is still nascent, the goal of this paper is to provide an empirical landscape of this field through a systematic literature review. A systematic literature search and screening process were considered and a final sample of 40 articles were included. First, this review presents the research contexts of the included studies such as geographical location, sample characteristics, data collection method, and the like. Second, it identifies drivers of social media fatigue and categorizes them into three conceptual levels: individual, relational, and environmental. Results of this systematic review outline several key directions for future research. Also, it contributes to a more nuanced theoretical understanding of how social media fatigue arises. Finally, identifying factors influencing fatigue helps social media service providers and health professionals to propose relevant intervention strategies to mitigate this phenomenon.