Leveraging social media networks for classification

Social media has reshaped the way in which people interact with each other. The rapid development of participatory web and social networking sites like YouTube, Twitter, and Facebook, also brings about many data mining opportunities and novel challenges. In particular, we focus on classification tasks with user interaction information in a social network. Networks in social media are heterogeneous, consisting of various relations. Since the relation-type information may not be available in social media, most existing approaches treat these inhomogeneous connections homogeneously, leading to an unsatisfactory classification performance. In order to handle the network heterogeneity, we propose the concept of social dimension to represent actors?? latent affiliations, and develop a classification framework based on that. The proposed framework, SocioDim, first extracts social dimensions based on the network structure to accurately capture prominent interaction patterns between actors, then learns a discriminative classifier to select relevant social dimensions. SocioDim, by differentiating different types of network connections, outperforms existing representative methods of classification in social media, and offers a simple yet effective approach to integrating two types of seemingly orthogonal information: the network of actors and their attributes.     

Mining tag-clouds to improve social media recommendation

Multimedia Tools and Applications - Massive amounts of data are available on social websites, therefore finding the suitable item is a challenging issue. According to recent social statistics, we...     

Detecting adverse drug reactions from social media based on multi-channel convolutional neural networks

Neural Computing and Applications - As one of the most important medical field subjects, adverse drug reaction seriously affects the patient’s life, health, and safety. Although many methods...     

A temporal graph framework for intelligence extraction in social media networks

Social media networks (SMNs) are increasingly used in professional management of knowledge workers and related assets. However, the factors affecting behavioral trends and activity levels in these networks are not well understood. Although social and cognitive theories can help to explain human behavior in traditional social networks, their application to SMNs has not been validated. Traditional social network modeling techniques may not accurately predict real-world SMN activities. This research developed a temporal graph framework for intelligence extraction in SMNs. Theory-based, data-driven models (Conformity Model (COM), Recency-Primacy Model (REM), Trend Interaction Model (TIM), Periodic Interaction Model (PIM)) were developed based on the framework to capture various aspects of user behavior: conformity effect, recency, primacy, periodicity, and dynamic trend. The models capture the activity history and dynamically combine pricing information to enhance predictive accuracy. Using data of 83,536 GitHub software repositories on cryptocurrency, this article reports the results of experiments that compare the models’ performance in predicting SMN activities over time. Experimental results show that the model (REM) that captures recency/primacy effects of human cognitive processing outperformed other models in 9 (out of 18) measures pertaining to engagement, contribution, influence, and popularity. Primacy plays a dominant role in predicting engagement, contribution, and popularity, whereas recency plays a key role in predicting influence. Short-term trend (modeled with TIM) was found to yield significantly better performance on predicting user contribution. The models also outperformed an integrated machine learning (IML) model by most measures. Overall, the effects modeled by REM and TIM were found to be more significant than the effects modeled by COM, PIM, and IML. The research contributes to enhancing understanding of SMN behavior, developing new models to simulate and predict SMN activities, and designing new artifacts for information systems practitioners to manage knowledge assets and to extract SMN intelligence.     

Recognizing CEO personality and its impact on business performance: Mining linguistic cues from social media

Upper echelons theory suggests that CEO personality will influence organizational performance. However, difficulty in measuring CEO personality restrains related research. We capture linguistic cues CEOs leaving on social media and recognize their personality by text mining. To our knowledge, it is the first study introducing social media text mining approaches into the research stream that empirically inquires and extends upper echelons theory. Then, we investigate the CEO personality’s impact on both operational and financial performance. Results show that CEO Extraversion, Emotional Stability, and Agreeableness improve Cost Efficiency and Profitability, while CEO Conscientiousness reduces them. CEO Openness to Experience negatively influences Profitability, and all facets of CEO personality improve Employee Productivity except for CEO Conscientiousness. The contribution of our research is multi-sided: (1). methodologically, we introduce a text mining approach to measure CEO personality; (2). theoretically, we provide empirical evidence for upper echelons theory; (3). practically, our results help companies evaluate CEO candidates from a personality perspective.     

Utopian social media

Inspired by the work on utopia of Ruth Levitas, and drawing on narrative inquiry and arts-based research methods, this paper presents the result of a series of social media utopia-making workshops where regular users of social media were invited to discuss the things that bother them and propose alternatives to the visions of control and manipulation that fill our reality and the research horizon. The paper is a combination of academic/methodological discourse and creative writing fuelled by the workshop results.     

Information sharing on social media sites

This study reports on an exploratory survey conducted to investigate the use of social media technologies for sharing information. This paper explores the issue of credibility of the information shared in the context of computer-mediated communication. Four categories of information were explored: sensitive, sensational, political and casual information, across five popular social media technologies: social networking sites, micro-blogging sites, wikis, online forums, and online blogs. One hundred and fourteen active users of social media technologies participated in the study. The exploratory analysis conducted in this study revealed that information producers use different cues to indicate credibility of the information they share on different social media sites. Organizations can leverage findings from this study to improve targeted engagement with their customers. The operationalization of how information credibility is codified by information producers contributes to knowledge in social media research.     

Detecting and responding to hostile disinformation activities on social media using machine learning and deep neural networks

Neural Computing and Applications - Disinformation attacks that make use of social media platforms, e.g., the attacks orchestrated by the Russian “Internet Research Agency” during the...     

复杂网络中的在线社交网络演化模型

在线社交网络是一种广泛存在的社会网络,其节点度遵循幂率分布规律,但对于其结构演化模型方面的相关研究还不多。基于复杂网络理论研究在线社交网络内部结构特征,提出一种结合内增长、外增长及内部边更替的演化模型,借助平均场理论分析该模型的拓扑特性,实验和理论分析表明由该模型生成的网络,其度分布服从幂率分布,且通过调整参数,幂率指数在1~3,能较好地反映不同类型的真实在线社交网络的度分布特征,因此具有广泛适用性。     

The diversity of why: a meta-analytical study of usage motivation in enterprise social networks

In times of demographic change, skill shortage and disruptive innovations, organizational knowledge management and innovative capacity are the key to a company’s success. But how can knowledge be retained with fast staff turnover, global project-based work and parental leaves? Using enterprise social media to improve knowledge dissemination at work seems promising, when looking at the success of private social networking sites. In this article we combine ten different empirical studies which investigated different aspects of how user diversity influences the motivation to use social media at work. The emerging meta-study using the DerSimonian–Laird method (total sample size \(N=522\)) analyzes different aspects of user diversity and their correlation with eight motives for SNS usage: information, importance, contact, self-presentation, autonomy, social comparison, and power and control. We found that that the individual achievement motivation correlates positively with the motives importance, power, information and self-presentation. The need for autonomy correlates with openness to new experiences and the need for social comparison with gender and neuroticism. From our findings, we derive practical implications for designing a social networking site for work which fulfills the users’ needs and functions along their motivation.     

Drawing on mobile crowds via social media

Recent years have witnessed the impact of crowdsourcing model, social media, and pervasive computing. We believe that the more significant impact is latent in the convergence of these ideas on the mobile platform. In this paper, we introduce a mobile crowdsourcing platform that is built on top of social media. A mobile crowdsourcing application called UbiAsk is presented as one study case. UbiAsk is designed for assisting foreign visitors by involving the local crowd to answer their image-based questions at hand in a timely fashion. Existing social media platforms are used to rapidly allocate microtasks to a wide network of local residents. The resulting data are visualized using a mapping tool as well as augmented reality (AR) technology, result in a visual information pool for public use. We ran a controlled field experiment in Japan for 6 weeks with 55 participants. The results demonstrated a reliable performance on response speed and response quantity: half of the requests were answered within 10 min, 75% of requests were answered within 30 min, and on average every request had 4.2 answers. Especially in the afternoon, evening and night, nearly 88% requests were answered in average approximately 10 min, with more than 4 answers per request. In terms of participation motivation, we found the top active crowdworkers were more driven by intrinsic motivations rather than any of the extrinsic incentives (game-based incentives and social incentives) we designed.     

A study of homophily on social media

The fact that similarity breeds connections, the principle of homophily, has been well-studied in existing sociology literature. Several studies have observed this phenomenon by conducting surveys on human subjects. These studies have concluded that new ties are formed between similar individuals. This phenomenon has been used to explain several socio-psychological concepts such as, segregation, community development, social mobility, etc. However, due to the nature of these studies and limitations because of involvement of human subjects, conclusions from these studies are not easily extensible in online social media. Social media, which is becoming the infinite space for interactions, has exceeded all the expectations in terms of growth, for reasons beyond human mind. New ties are formed in social media in the same way that they emerge in real-world. However, given the differences between real world and online social media, do the same factors that govern the construction of new ties in real world also govern the construction of new ties in social media? In other words, does homophily exist in social media? In this article, we study this extremely significant question. We propose a systematic approach by studying three online social media sites, BlogCatalog, Last.fm, and LiveJournal and report our findings along with some interesting observations. The results indicate that the influence of interest-based homophily is not a very strong leading factor for constructing new ties specifically in the three social media sites with implications to strategic advertising, recommendations, and promoting applications at large.     

Sockpuppet gang detection on social media sites

Users of social media sites can use more than one account. These identities have pseudo anonymous properties, and as such some users abuse multiple accounts to perform undesirable actions, such as posting false or misleading remarks comments that praise or defame the work of others. The detection of multiple user accounts that are controlled by an individual or organization is important. Herein, we define the problem as sockpuppet gang (SPG) detection. First, we analyze user sentiment orientation to topics based on emotional phrases extracted from their posted comments. Then we evaluate the similarity between sentiment orientations of user account pairs, and build a similar-orientation network (SON) where each vertex represents a user account on a social media site. In an SON, an edge exists only if the two user accounts have similar sentiment orientations to most topics. The boundary between detected SPGs may be indistinct, thus by analyzing account posting behavior features we propose a multiple random walk method to iteratively remeasure the weight of each edge. Finally, we adopt multiple community detection algorithms to detect SPGs in the network. User accounts in the same SPG are considered to be controlled by the same individual or organization. In our experiments on real world datasets, our method shows better performance than other contemporary methods.     

Deep sentiment analysis: Mining the causality between personality-value-attitude for analyzing business ads in social media

IT vendors routinely use social media such as YouTube not only to disseminate their IT product information, but also to acquire customer input efficiently as part of their market research strategies. Customer responses that appear in social media, however, are typically unstructured; thus, a fairly large data set is needed for meaningful analysis. Although identifying customers’ value structures and attitudes may be useful for developing targeted or niche markets, the unstructured and volume-heavy nature of customer data prohibits efficient and economical extraction of such information. Automatic extraction of customer information would be valuable in determining value structure and strength. This paper proposes an intelligent method of estimating causality between user profiles, value structures, and attitudes based on the replies and published content managed by open social network systems such as YouTube. To show the feasibility of the idea proposed in this paper, information richness and agility are used as underlying concepts to create performance measures based on media/information richness theory. The resulting deep sentiment analysis proves to be superior to legacy sentiment analysis tools for estimation of causality among the focal parameters.     

Improving media services on P2P networks

The Media Accelerating Peer Services system extends P2P infrastructures to improve multimedia services across heterogeneous computing platforms. In this article, we present an architecture and resource management and adaptation framework that transcends existing infrastructures to accommodate and accelerate multimedia peer applications and services. We also propose key technology components that support seamless adaptation of resources to enhance quality of service and the building of better tools and applications that utilize the peer-computing network's underlying power     

Characteristics of social media stories

An emerging trend in social media is for users to create and publish “stories”, or curated lists of  Web resources, with the purpose of creating a particular narrative of interest to the user. While some stories on the Web are automatically generated, such as Facebook’s “Year in Review”, one of the most popular storytelling services is “Storify”, which provides users with curation tools to select, arrange, and annotate stories with content from social media and the Web at large. We would like to use tools, such as Storify, to present (semi-)automatically created summaries of archival collections. To support automatic story creation, we need to better understand as a baseline the structural characteristics of popular (i.e., receiving the most views) human-generated stories. We investigated 14,568 stories from Storify, comprising 1,251,160 individual resources, and found that popular stories (i.e., top 25 % of views normalized by time available on the Web) have the following characteristics: 2/28/1950 elements (min/median/max), a median of 12 multimedia resources (e.g., images, video), 38 % receive continuing edits, and 11 % of their elements are missing from the live Web. We also checked the population of Archive-It collections (3109 collections comprising 305,522 seed URIs) for better understanding the characteristics of the collections that we intend to summarize. We found that the resources in human-generated stories are different from the resources in Archive-It collections. In summarizing a collection, we can only choose from what is archived (e.g., twitter.com is popular in Storify, but rare in Archive-It). However, some other characteristics of human-generated stories will be applicable, such as the number of resources.     

Inferring urban air quality based on social media

Outdoor air pollution is a serious environmental problem in many developing countries; obtaining timely and accurate information about urban air quality is a first step toward air pollution control. Many developing countries however, do not have any monitoring stations and therefore the means to measure air quality. We address this problem by using social media to collect urban air quality information and propose a method for inferring urban air quality in Chinese cities based on China's largest social media platform, Sina Weibo combined with other meteorological data. Our method includes a data crawler to locate and acquire air-quality associated historical Weibo data, a procedure for extracting indicators from these Weibo and factors from meteorological data, a model to infer air quality index (AQI) of a city based on the extracted Weibo indicators supported by meteorological factors. We implemented the proposed method in case studies at Beijing, Shanghai, and Wuhan, China. The results show that based the Weibo indicators and meteorological factors we extracted, this method can infer the air quality conditions of a city within narrow margins of error. The method presented in this article can aid air quality assessment in cities with few or even no air quality monitoring stations.