Mining user interests over active topics on social networks

Inferring users’ interests from their activities on social networks has been an emerging research topic in the recent years. Most existing approaches heavily rely on the explicit contributions (posts) of a user and overlook users’ implicit interests, i.e., those potential user interests that the user did not explicitly mention but might have interest in. Given a set of active topics present in a social network in a specified time interval, our goal is to build an interest profile for a user over these topics by considering both explicit and implicit interests of the user. The reason for this is that the interests of free-riders and cold start users who constitute a large majority of social network users, cannot be directly identified from their explicit contributions to the social network. Specifically, to infer users’ implicit interests, we propose a graph-based link prediction schema that operates over a representation model consisting of three types of information: user explicit contributions to topics, relationships between users, and the relatedness between topics. Through extensive experiments on different variants of our representation model and considering both homogeneous and heterogeneous link prediction, we investigate how topic relatedness and users’ homophily relation impact the quality of inferring users’ implicit interests. Comparison with state-of-the-art baselines on a real-world Twitter dataset demonstrates the effectiveness of our model in inferring users’ interests in terms of perplexity and in the context of retweet prediction application. Moreover, we further show that the impact of our work is especially meaningful when considered in case of free-riders and cold start users.     

Towards a big data framework for analyzing social media content

Modern companies generate value by digitalizing their services and products. Knowing what customers are saying about the firm through reviews in social media content constitutes a key factor to succeed in the big data era. However, social media data analysis is a complex discipline due to the subjectivity in text review and the additional features in raw data. Some frameworks proposed in the existing literature involve many steps that thereby increase their complexity. A two-stage framework to tackle this problem is proposed: the first stage is focused on data preparation and finding an optimal machine learning model for this data; the second stage relies on established layers of big data architectures focused on getting an outcome of data by taking most of the machine learning model of stage one. Thus, a first stage is proposed to analyze big and small datasets in a non-big data environment, whereas the second stage analyzes big datasets by applying the first stage machine learning model of. Then, a study case is presented for the first stage of the framework to analyze reviews of hotel-related businesses. Several machine learning algorithms were trained for two, three and five classes, with the best results being found for binary classification.     

Towards a real-time processing framework based on improved distributed recurrent neural network variants with fastText for social big data analytics

Big data generated by social media stands for a valuable source of information, which offers an excellent opportunity to mine valuable insights. Particularly, User-generated contents such as reviews, recommendations, and users’ behavior data are useful for supporting several marketing activities of many companies. Knowing what users are saying about the products they bought or the services they used through reviews in social media represents a key factor for making decisions. Sentiment analysis is one of the fundamental tasks in Natural Language Processing. Although deep learning for sentiment analysis has achieved great success and allowed several firms to analyze and extract relevant information from their textual data, but as the volume of data grows, a model that runs in a traditional environment cannot be effective, which implies the importance of efficient distributed deep learning models for social Big Data analytics. Besides, it is known that social media analysis is a complex process, which involves a set of complex tasks. Therefore, it is important to address the challenges and issues of social big data analytics and enhance the performance of deep learning techniques in terms of classification accuracy to obtain better decisions.In this paper, we propose an approach for sentiment analysis, which is devoted to adopting fastText with Recurrent neural network variants to represent textual data efficiently. Then, it employs the new representations to perform the classification task. Its main objective is to enhance the performance of well-known Recurrent Neural Network (RNN) variants in terms of classification accuracy and handle large scale data. In addition, we propose a distributed intelligent system for real-time social big data analytics. It is designed to ingest, store, process, index, and visualize the huge amount of information in real-time. The proposed system adopts distributed machine learning with our proposed method for enhancing decision-making processes. Extensive experiments conducted on two benchmark data sets demonstrate that our proposal for sentiment analysis outperforms well-known distributed recurrent neural network variants (i.e., Long Short-Term Memory (LSTM), Bidirectional Long Short-Term Memory (BiLSTM), and Gated Recurrent Unit (GRU)). Specifically, we tested the efficiency of our approach using the three different deep learning models. The results show that our proposed approach is able to enhance the performance of the three models. The current work can provide several benefits for researchers and practitioners who want to collect, handle, analyze and visualize several sources of information in real-time. Also, it can contribute to a better understanding of public opinion and user behaviors using our proposed system with the improved variants of the most powerful distributed deep learning and machine learning algorithms. Furthermore, it is able to increase the classification accuracy of several existing works based on RNN models for sentiment analysis.     

Exploring computer supported collaborative coordination through social networks

The performance of work groups and in particular geographically distributed ones is negatively affected by communication issues and task dependencies.Contemporary science suggests social link optimization apart from improving the technical aspects to address these issues. In our study, we focus on distributed coordination and project performance. Social network structure and coordination performance variables are described by our framework with regards to distributed coordination during bug fixing process. Based on the model and the literature reviewed, we propose two propositions—(i) the level of interconnectedness has a negative relation with coordination performance; and (ii) centrality social network measures have positive relation with coordination performance variables. We use a sample of 415 Open Source Projects hosted on SourceForge.net. The results suggest that both propositions are correct. Furthermore, in the methods section implementation of an automated process is introduced to build graph definitions in adjacency matrix or NCOL format from thousands of forum threads. We describe the implementation of a novel method to plot sociograms in batch from hundreds of graph definitions automatically and calculate network centrality and density measures for all of them at the same time. Finally, we suggest the implications of this study to software development project management research.     

Co-authorship networks and research impact: A social capital perspective

The impact of research work is related to a scholar's reputation and future promotions. Greater research impact not only inspires scholars to continue their research, but also increases the possibility of a larger research budget from sponsors. Given the importance of research impact, this study proposes that utilizing social capital embedded in a social structure is an effective way to achieve more research impact. The contribution of this study is to define six indicators of social capital (degree centrality, closeness centrality, betweenness centrality, prolific co-author count, team exploration, and publishing tenure) and investigate how these indicators interact and affect citations for publications. A total of 137 Information Systems scholars from the Social Science Citation Index database were selected to test the hypothesized relationships. The results show that betweenness centrality plays the most important role in taking advantage of non-redundant resources in a co-authorship network, thereby significantly affecting citations for publications. In addition, we found that prolific co-author count, team exploration, and publishing tenure all have indirect effects on citation count. Specifically, co-authoring with prolific scholars helps researchers develop centralities and, in turn, generate higher numbers of citations. Researchers with longer publishing tenure tend to have higher degree centrality. When they collaborate more with different scholars, they achieve more closeness and betweenness centralities, but risk being distrusted by prolific scholars and losing chances to co-author with them. Finally, implications of findings and recommendations for future research are discussed.     

A survey towards an integration of big data analytics to big insights for value-creation

Big Data Analytics (BDA) is increasingly becoming a trending practice that generates an enormous amount of data and provides a new opportunity that is helpful in relevant decision-making. The developments in Big Data Analytics provide a new paradigm and solutions for big data sources, storage, and advanced analytics. The BDA provide a nuanced view of big data development, and insights on how it can truly create value for firm and customer. This article presents a comprehensive, well-informed examination, and realistic analysis of deploying big data analytics successfully in companies. It provides an overview of the architecture of BDA including six components, namely: (i) data generation, (ii) data acquisition, (iii) data storage, (iv) advanced data analytics, (v) data visualization, and (vi) decision-making for value-creation. In this paper, seven V's characteristics of BDA namely Volume, Velocity, Variety, Valence, Veracity, Variability, and Value are explored. The various big data analytics tools, techniques and technologies have been described. Furthermore, it presents a methodical analysis for the usage of Big Data Analytics in various applications such as agriculture, healthcare, cyber security, and smart city. This paper also highlights the previous research, challenges, current status, and future directions of big data analytics for various application platforms. This overview highlights three issues, namely (i) concepts, characteristics and processing paradigms of Big Data Analytics; (ii) the state-of-the-art framework for decision-making in BDA for companies to insight value-creation; and (iii) the current challenges of Big Data Analytics as well as possible future directions.     

Privacy threat model for data portability in social network applications

The advent of the participatory Web and social network applications has changed our communication behaviour and the way we express ourselves on the Web. Social network application providers benefit from the increasing amount of personally identifiable information willingly displayed on their sites but, at the same time, risks of data misuse threaten the information privacy of individual users as well as the providers’ business model. From recent research, this paper reports the major requirements for developing privacy-preserving social network applications and proposes a privacy threat model that can be used to enhance the information privacy in data or social network portability initiatives by determining the issues at stake related to the processing of personally identifiable information.     

Using big data analytics to study brand authenticity sentiments: The case of Starbucks on Twitter

There is a strong interest among academics and practitioners in studying branding issues in the big data era. In this article, we examine the sentiments toward a brand, via brand authenticity, to identify the reasons for positive or negative sentiments on social media. Moreover, in order to increase precision, we investigate sentiment polarity on a five-point scale. From a database containing 2,282,912 English tweets with the keyword ‘Starbucks’, we use a set of 2204 coded tweets both for analyzing brand authenticity and sentiment polarity. First, we examine the tweets qualitatively to gain insights about brand authenticity sentiments. Then we analyze the data quantitatively to establish a framework in which we predict both the brand authenticity dimensions and their sentiment polarity. Through three qualitative studies, we discuss several tweets from the dataset that can be classified under the quality commitment, heritage, uniqueness, and symbolism categories. Using latent semantic analysis (LSA), we extract the common words in each category. We verify the robustness of previous findings with an in-lab experiment. Results from the support vector machine (SVM), as the quantitative research method, illustrate the effectiveness of the proposed procedure of brand authenticity sentiment analysis. It shows high accuracy for both the brand authenticity dimensions’ predictions and their sentiment polarity. We then discuss the theoretical and managerial implications of the studies.     

Co-citation and cluster analyses of extant literature on social networks

Over one billion people are currently using social media such as social websites (Facebook Newsroom, 2015); consequently, numerous academic scholars have developed interest in studying the use of social media and social networks. However, few studies have focused on examining the core factors of social networks. In this study, we collected studies on social-network-related topics that were published between January 1996 and December 2014, assembling a total of 2565 articles and 81,316 citations. Co-citation analysis and cluster analysis were applied to verify seven main factors regarding social networks: (a) the measure of complex social networks; (b) community structure; (c) strong and weak ties; (d) the evolution of social networks; (e) network structure and relationship; (f) value concept and measurement strategies; and (g) social capital. Finally, the results of this study were further discussed to elucidate the core topics relevant to social networks.     

Actualizing big data analytics for smart cities: A cascading affordance study

This study investigates how Big Data Analytics (BDA) can be leveraged to support a city’s transformation into a smart destination. We conduct an in-depth case study of a city-in-transformation and adopt the perspective of technology affordances to uncover the varying opportunities enabled by BDA to facilitate the attainment of smart tourism goals. Our findings unveil three types of BDA affordances and demonstrate how these affordances are actualized in a cascading manner to enable informed decisions and a sustainable development of smart tourism. Implications are presented for future investigation of the affordances of BDA in smart tourism, as well as for policy makers and practitioners who engage in the development of innovative tourism services for the smart citizens.     

A big data analytics model for customer churn prediction in the retiree segment

Undoubtedly, the change in consumers’ choices and expectations, stemming from the emerging technology and also significant availability of different products and services, created a highly competitive landscape in various customer service sectors, including the financial industry. Accordingly, the Canadian banking industry has also become highly competitive due to the threats and disruptions caused by not only direct competitors, but also new entrants to the market.The primary objective of this paper is to construct a predictive churn model by utilizing big data, including the structured archival data, integrated with unstructured data from sources such as online web pages, the number of website visits and phone conversation logs, for the first time in the financial industry. It also examines the effect of different aspects of customers’ behavior on churning decisions. The Datameer big data analytics tool on the Hadoop platform and predictive techniques using the SAS business intelligence system were applied to study the client retirement journey path and to create a churn prediction model. By deploying the above systems, we were able to uncover a wealth of data and information associated with over 3 million customers’ records within the retiree segment of the target bank, from 2011 to 2015.     

FALCoN: Detecting and classifying abusive language in social networks using context features and unlabeled data

Social networks have grown into a widespread form of communication that allows a large number of users to participate in conversations and consume information at any time. The casual nature of social media allows for nonstandard terminology, some of which may be considered rude and derogatory. As a result, a significant portion of social media users is found to express disrespectful language. This problem may intensify in certain developing countries where young children are granted unsupervised access to social media platforms. Furthermore, the sheer amount of social media data generated daily by millions of users makes it impractical for humans to monitor and regulate inappropriate content. If adolescents are exposed to these harmful language patterns without adequate supervision, they may feel obliged to adopt them. In addition, unrestricted aggression in online forums may result in cyberbullying and other dreadful occurrences. While computational linguistics research has addressed the difficulty of detecting abusive dialogues, issues remain unanswered for low-resource languages with little annotated data, leading the majority of supervised techniques to perform poorly. In addition, social media content is often presented in complex, context-rich formats that encourage creative user involvement. Therefore, we propose to improve the performance of abusive language detection and classification in a low-resource setting, using both the abundant unlabeled data and the context features via the co-training protocol that enables two machine learning models, each learning from an orthogonal set of features, to teach each other, resulting in an overall performance improvement. Empirical results reveal that our proposed framework achieves F1 values of 0.922 and 0.827, surpassing the state-of-the-art baselines by 3.32% and 45.85% for binary and fine-grained classification tasks, respectively. In addition to proving the efficacy of co-training in a low-resource situation for abusive language detection and classification tasks, the findings shed light on several opportunities to use unlabeled data and contextual characteristics of social networks in a variety of social computing applications.     

A framework for understanding online group behaviors during a catastrophic event

This study investigated the underlying mechanisms of online social media group behaviors in an emergency. The proposed framework was designed to analyze group behaviors/interactions and examine the main topics of interest among numerous tweets generated in an emergency. We collected tweets sent during Hurricane Harvey in 2017 and applied the framework to demonstrate its effectiveness. The proposed framework enables us to understand the unique characteristics of group interactions and develop operational strategies to effectively communicate with the public, as well as other groups, as critical emergency information appears in an online social network.     

An approach for planning and deploying gamification concepts with social networks within educational contexts

Gamification planning has been a topic of discussion in the last years since it can be used to increase performance, engagement, and motivation of end users. When properly applied in educational settings, gamification can lead to better learning. Furthermore, it can be boosted when tied to social networks. However, according to the literature, there are three main concerns regarding this topic: (a) instructors and teachers does not have the resources to plan and develop gamification strategies into their classes; (b) gamification needs a systematic approach to achieve the desired positive results; and (c) inexistence of systematic approaches that connect and help in the design of gamification and social network tasks within these contexts. Thus, this work proposes a solution to help instructors and teachers to plan and deploy gamification concepts with social network features in learning environments. In this paper, we detailed our approach depicting the set of items to analyze and compare it with other solutions that are focused on education. Then, it was conducted a case study over a programming course (N = 40) to analyze the planning and deployment phases. Our results demonstrated that our approach is the first to consider the stakeholders (i.e. instructors and teachers) as part of the process. Moreover, even though there are still some obstacles to overcome, the gamified strategies that were created achieved positive acceptance among the students and professor.     

A health data analytics maturity model for hospitals information systems

In the last five decades, maturity models have been introduced as reference frameworks for Information System (IS) management in organizations within different industries. In the healthcare domain, maturity models have also been used to address a wide variety of challenges and the high demand for hospital IS (HIS) implementations. The increasing volume of data, is exceeded the ability of health organizations to process it for improving clinical and financial efficiencies and quality of care. It is believed that careful and attentive use of Data Analytics in healthcare can transform data into knowledge that can improve patient outcomes and operational efficiency. A maturity model in this conjuncture, is a way of identifying strengths and weaknesses of the HIS maturity and thus, find a way for improvement and evolution. This paper presents a proposal to measure Hospitals Information Systems maturity with regard to Data Analytics. The outcome of this paper is a maturity model, which includes six stages of HIS growth and maturity progression.     

Real-time big data processing for anomaly detection: A Survey

The advent of connected devices and omnipresence of Internet have paved way for intruders to attack networks, which leads to cyber-attack, financial loss, information theft in healthcare, and cyber war. Hence, network security analytics has become an important area of concern and has gained intensive attention among researchers, off late, specifically in the domain of anomaly detection in network, which is considered crucial for network security. However, preliminary investigations have revealed that the existing approaches to detect anomalies in network are not effective enough, particularly to detect them in real time. The reason for the inefficacy of current approaches is mainly due the amassment of massive volumes of data though the connected devices. Therefore, it is crucial to propose a framework that effectively handles real time big data processing and detect anomalies in networks. In this regard, this paper attempts to address the issue of detecting anomalies in real time. Respectively, this paper has surveyed the state-of-the-art real-time big data processing technologies related to anomaly detection and the vital characteristics of associated machine learning algorithms. This paper begins with the explanation of essential contexts and taxonomy of real-time big data processing, anomalous detection, and machine learning algorithms, followed by the review of big data processing technologies. Finally, the identified research challenges of real-time big data processing in anomaly detection are discussed.     

A knowledge management approach to capture organizational learning networks

Effective knowledge management practices in organizations are focused on knowledge creation and knowledge transfer activities. Thus, intelligence and competencies matters at the organizational workplace. For most knowledge intensive organizations is fundamental the continuous availability and development of domain expertise. This paper describes an ongoing research project to develop an organizational knowledge architecture that is being specified and developed to support collaboration tasks as well as design and model predictive data analysis and insights for organizational development. The primary goal of this research is to create a suitable architecture for use, initially, in intranet (corporate portal) collaborative procedures, but also scalable for later use in more generic forms of ontology-driven knowledge management systems. The designed architecture and functionalities aim to create coherent web data layers for intranet learning and predictive analysis, defining the vocabulary and semantics for knowledge sharing and reuse projects. Regarding intellectual capital definition, this research argues that effective knowledge management are based on the dynamic nature of the organizational knowledge, and predictive data analysis and insights identification can transform and add value to an organization. This paper presents a knowledge management and engineering perspective (ontology based) for the application of predictive analysis and insights at the organizational (corporate) workplace towards the development of the organizational learning network.     

中国大数据政策体系演化研究

在大数据快速发展的背景下，制定科学、合理的大数据政策体系，支撑大数据高质量发展变得日益迫切。本文通过构建“政策工具-政策主题词”二维分析框架，对2000年以来国家层面有关大数据发展的政策文本进行了分析，从大数据政策的演进脉络、政策关系网络、政策主题词演进态势以及政策工具的使用情况等四个方面探究了大数据政策文本内容。研究结果表明：我国大数据政策发展，大致经历了从强调基础设施建设到关注产业培育与创新，再到构建大数据政策体系的发展过程。在发展的不同阶段，政策主题词与当时的经济、社会、国际环境密切相关。目前我国已经初步建立起大数据政策网络体系，但整体还处于初步发展阶段，大数据政策的顶层设计还不健全，现有政策工具中知识产权保护类政策工具和政府采购类政策工具使用较少等，仍需要进一步加强完善。     

A multilevel model for effects of social capital and knowledge sharing in knowledge-intensive work teams

Although it is a widely held belief that social capital facilitates knowledge sharing among individuals, there is little research that has deeply investigated the impacts of social capital at different levels on an individual's knowledge sharing behavior. To address this research gap, this study combines a multilevel approach and an optimal network configuration view to investigate the multilevel effects of social capital on individuals’ knowledge sharing in knowledge intensive work teams. This study makes a distinction between the social capital at the team-level and that of social capital at the individual level to examine their cross-level and direct effects on an individual's sharing of explicit and tacit knowledge. A survey involving 343 participants in 47 knowledge-intensive teams was conducted for testing the multilevel model. The results reveal that social capital at both levels jointly influences an individual's explicit and tacit knowledge sharing. Further, when individuals possess a moderate betweenness centrality and the whole team holds a moderate network density, team members’ knowledge sharing can be maximized. These findings offer a more comprehensive and precise understanding of the multilevel impacts of social capital on team members’ knowledge sharing behavior, thus contributing to the social capital theory, as well as knowledge management research and practices.