Search

Community detection is a critical component of network analysis and a hot topic in social computing. Detecting community structure in dynamic networks has important theoretical and practical implications for understanding the intrinsic function of networks and predicting network behavior. However, the majority of existing dynamic community detection methods adopt shallow models, which have limited ability to excavate complex non-linear structures and tend to generate undesirable community structures. In order to obtain an accurate and robust community structure in dynamic networks, we are inspired by network representation learning and utilize the deep learning to detect evolving communities in dynamic networks. In this paper, we propose a novel dynamic community detection method by fusing Deep Learning and Evolutionary Clustering (DLEC). This work attempts to combine deep learning and evolutionary clustering into a unified framework. First, we propose a matrix construction strategy to fully reveal the inherent community structures via the underlying community memberships. Then, we develop a novel multi-layer deep autoencoder framework that consists of multiple non-linear functions to extract the latent deep representation of the dynamic network. Based on the evolutionary clustering framework, a graph regularization term is introduced to ensure the smoothness of the community evolution. Finally, we employ the K-means clustering algorithm on the low-dimensional network space to obtain the community structure. Extensive experimental results on synthetic and real-world networks show that the proposed DLEC algorithm can effectively detect high-quality communities in dynamic networks. [ABSTRACT FROM AUTHOR]

We are in the early stages of a new era of demographic research that offers exciting opportunities to quantify demographic phenomena at a scale and resolution once unimaginable. These scientific possibilities are opened up by new sources of data, such as the digital traces that arise from ubiquitous social computing, massive longitudinal datasets produced by the digitization of historical records, and information about previously inaccessible populations reached through innovations in classic modes of data collection. In this commentary, we describe five promising new sources of demographic data and their potential appeal. We identify cross‐cutting challenges shared by these new data sources and argue that realizing their full potential will demand both innovative methodological developments and continued investment in high‐quality, traditional surveys and censuses. Despite these considerable challenges, the future is bright: these new sources of data will lead demographers to develop new theories and revisit and sharpen old ones. [ABSTRACT FROM AUTHOR]

As virtual teamwork became widespread, the importance of the usability of groupware has increased, as groupware must have high usability in order to properly support teamwork. Several studies have focused on the usability evaluation of groupware, yet, both in terms of science and practice, there is a lack of a method that, like existing empirical single-user methods, can be routinely applied in software development practice. The new Team Usability Testing method differs from previous methods in that it is an empirical method for evaluating synchronous (real-time) collaborative software that explores the team usability problems with the help of real or potential users. The method consists of questionnaires, screen recording videos and group interviews, and the framework of data processing is based on the theory of the mechanics of collaboration. In the four steps of the development of the method, empirical studies under laboratory conditions and real working conditions, and then, for validation purposes, analytical, heuristic evaluation with the involvement of usability experts were performed. During the development and the application of the method, the communication patterns of different teams were examined in terms of their relationship to software usability. In this paper we present the results of the second laboratory study and the heuristic evaluation in details. Overall, the method is able to explore team usability problems of different types of teams, making it a valuable element in existing methods for evaluating the usability of groupware. [ABSTRACT FROM AUTHOR]

Web annotation environments are widely used in education based on the premise that student interaction in these environments benefits individual and group learning. However, there is little research on factors driving student interaction in web annotation activities. In this study we asked: What dynamics could explain social interaction among students in web annotation activities in college classrooms? Recognizing the mediated nature of social interaction in web annotation, we hypothesized that student interaction is driven by multiple factors including previous relational events and semantic features of annotation content. Following a novel network analysis method named relational event modeling, we analyzed a rich dataset from four online classes. Results indicated annotation popularity was initially predictive of student replies, meaning popular annotation threads were more likely to attract new replies. However, this effect nearly diminished when adding thread-level semantic cohesion to the model, indicating an significant role played by semantic cohesion in attracting new responses. This paper makes important progress towards modeling social interaction in digital environments as a dynamic, mediated phenomenon. This study contributes empirical insights into web annotation and calls for future work to investigate social interaction as a dynamic network phenomenon. [ABSTRACT FROM AUTHOR]

Soft skills, such as personality characteristics, communication skills and leadership, affect personal career performance greatly. Therefore, predicting the soft skills of interviewees can provide interviewers with a strong reference for the decision of hiring. Nowadays, as asynchronous video interviews have gradually become a popular form of interviews, automatic interview evaluation of soft skills has attracted widespread attention from researchers. However, existing automatic evaluation methods have two significant drawbacks. First, most of them model the problem as multi-modal fusion of long-term sequences, while ignoring the consistency of multi-modal expression in short-time windows, which is a key attribute of the interview scene. Second, without embedding of professional knowledge in the interview field, the interpretability of the model is relatively weak. To address the above problems, we propose a novel Multi-modal Window-Consistency Fusion network, namely MWCF, to capture the expression consistency of different modalities in a short-time window and re-weight the language signals to enhance important portions in verbal clues. Meanwhile, in order to enhance the interpretability of the evaluation model, we introduce the professional knowledge of interviewers by proposing a topic generation module based on question attention, and embedding the most representative keywords under different soft skills into the model. Furthermore, a real-world interview dataset is built by developing an asynchronous interview platform, and extensive experiments are conducted to show the superior performance of our proposed model. [ABSTRACT FROM AUTHOR]

Stance detection is a crucial task in natural language processing and social computing, focusing on classifying expressed attitudes towards specific targets based on the input text. Conventional methods predominantly view stance detection as a task of target-oriented, sentence-level text classification. On popular social media platforms like Twitter, users often express their opinions through hashtags in addition to textual content within tweets. However, current methods primarily treat hashtags as data retrieval labels, neglecting to effectively utilize the semantic information they carry. In this paper, we propose a large language model knowledge-enhanced stance detection framework (LKESD) for stance detection. LKESD contains three main components: an instruction-prompted background knowledge acquisition module (IPBKA) that retrieves background knowledge of hashtags by providing handcrafted prompts to large language models (LLMs); a graph convolutional feature-enhancement module (GCFEM) is designed to extract the semantic representations of words that frequently co-occur with hashtags in the dataset by leveraging textual associations; an a knowledge fusion network (KFN) is proposed to selectively integrate graph representations and LLM features using a prompt-tuning framework. Extensive experimental results on three benchmark datasets demonstrate that our LKESD method outperforms 2.7% on all setups over compared methods, validating its effectiveness in stance detection tasks. [ABSTRACT FROM AUTHOR]

The proliferation of fake news on social media platforms necessitates the development of reliable datasets for effective fake news detection and veracity analysis. In this article, we introduce a veracity dataset of Arabic tweets called “VERA-ARAB”, a pioneering large-scale dataset designed to enhance fake news detection in Arabic tweets. VERA-ARAB is a balanced, multi-domain, and multi-dialectal dataset, containing both fake and true news, meticulously verified by fact-checking experts from Misbar. Comprising approximately 20,000 tweets from 13,000 distinct users and covering 884 different claims, the dataset includes detailed information such as news text, user details, and spatiotemporal data, spanning diverse domains like sports and politics. We leveraged the X API to retrieve and structure the dataset, providing a comprehensive data dictionary to describe the raw data and conducting a thorough statistical descriptive analysis. This analysis reveals insightful patterns and distributions, visualized according to data type and nature. We also evaluated the dataset using multiple machine learning classification models, exploring various social and textual features. Our findings indicate promising results, particularly with textual features, underscoring the dataset’s potential for enhancing fake news detection. Furthermore, we outline future work aimed at expanding VERA-ARAB to establish it as a benchmark for Arabic tweets in fake news detection. We also discuss other potential applications that could leverage the VERA-ARAB dataset, emphasizing its value and versatility for advancing the field of fake news detection in Arabic social media. Potential applications include user veracity assessment, topic modeling, and named entity recognition, demonstrating the dataset's wide-ranging utility for broader research in information quality management on social media.

Information and communication technologies enable the transformation of traditional energy systems into cyber-physical energy systems (CPESs), but such systems have also become popular targets of cyberattacks. Currently, available methods for evaluating the impacts of cyberattacks suffer from limited resilience, efficacy, and practical value. To mitigate their potentially disastrous consequences, this study suggests a two-stage, discrepancy-based optimization approach that considers both preparatory actions and response measures, integrating concepts from social computing. The proposed Kullback-Leibler divergence-based, distributionally robust optimization (KDR) method has a hierarchical, two-stage objective function that incorporates the operating costs of both system infrastructures (e.g., energy resources, reserve capacity) and real-time response measures (e.g., load shedding, demand-side management, electric vehicle charging station management). By incorporating social computing principles, the optimization framework can also capture the social behavior and interactions of energy consumers in response to cyberattacks. The preparatory stage entails day-ahead operational decisions, leveraging insights from social computing to model and predict the behaviors of individuals and communities affected by potential cyberattacks. The mitigation stage generates responses designed to contain the consequences of the attack by directing and optimizing energy use from the demand side, taking into account the social context and preferences of energy consumers, to ensure resilient, economically efficient CPES operations. Our method can determine optimal schemes in both stages, accounting for the social dimensions of the problem. An original disaster mitigation model uses an abstract formulation to develop a risk-neutral model that characterizes cyberattacks through KDR, incorporating social computing techniques to enhance the understanding and response to cyber threats. This approach can mitigate the impacts more effectively than several existing methods, even with limited data availability. To extend this risk-neutral model, we incorporate conditional value at risk as an essential risk measure, capturing the uncertainty and diverse impact scenarios arising from social computing factors. The empirical results affirm that the KDR method, which is enriched with social computing considerations, produces resilient, economically efficient solutions for managing the impacts of cyberattacks on a CPES. By integrating social computing principles into the optimization framework, it becomes possible to better anticipate and address the social and behavioral aspects associated with cyberattacks on CPESs, ultimately improving the overall resilience and effectiveness of the system's response measures.

Digitally enabled means for judgment aggregation have renewed interest in "wisdom of the crowd" effects and kick-started collective intelligence design as an emerging field in the cognitive and computational sciences. A keenly debated question here is whether social influence helps or hinders collective accuracy on estimation tasks, with recent results on the role of network structure hinting at a reconciliation of seemingly contradictory past results. Yet, despite a growing body of literature linking social network structure and collective accuracy, strategies for exploiting network structure to harness crowd wisdom are underexplored. We introduce one such strategy: rewiring algorithms that dynamically manipulate the structure of communicating social networks. Through agent-based simulations and an online multiplayer experiment, we provide a proof of concept showing how rewiring algorithms can increase the accuracy of collective estimations—even in the absence of knowledge of the ground truth. However, we also find that the algorithms' effects are contingent on the distribution of estimates initially held by individuals before communication occurs. CCS Concepts: •Human-centered computing → Collaborative and social computing• Applied computing → Psychology. [ABSTRACT FROM AUTHOR]

This document is a summary of a special issue titled "Artificial Intelligence in Complex Networks." The special issue explores the integration of artificial intelligence (AI) into complex networks and its impact on various scientific research areas. The articles cover topics such as key node identification, community detection, recommendation systems, object detection, data processing, and optimal decision algorithms in complex networks. The papers were contributed by researchers from Europe and Asia, showcasing a diverse range of perspectives and techniques. The document emphasizes the interdisciplinary nature of studying complex networks and highlights the need for responsible and ethical implementation of AI technologies. [Extracted from the article]

Smartphones have become highly personalized interfaces between people and the technology ecosystem around them. In this sense, they play a key role for a technology shift from the current Internet of Things to a future human-centric paradigm of an Internet of People, automatically adapting smart things and services to the preferences and context of their users. In this paper, we propose the use of Complex Event Processing (CEP) engines deployed in the users' smartphones granting them context-awareness capabilities in order to react to external stimulus, and enabling them to interact both with smart things and services in the surroundings of the users. With that purpose, we have designed a communication architecture that interconnects CEP engines running on smartphones, providing a framework for building applications for Mobile-based Collaborative Social Computing (MCSC). For that, we make use of previous works of the authors with Digital Avatars, a framework which promotes the use of smartphones for inferring and sharing a unique digital avatar or virtual profile of each user. The resulting framework, which we have called Collaborative CEP, allows to implement complex interactions among users, and between them and the IoT, a common need in Collaborative Social Computing applications. We provide a proof of concept based on the implementation of a Cops and Robbers game to test the expressiveness and correct functioning of the framework, and we evaluate its performance and efficiency. [ABSTRACT FROM AUTHOR]

Continuous reinforcement learning carries potential security risks when applied in real-world scenarios, which could have significant societal implications. While its field of application is expanding, the majority of applications still remain confined to virtual environments. If only a single continuous learning method is applied to an unmanned system, it will still forget previously learned experiences, and retraining will be required when it encounters unknown environments. This reduces the learning efficiency of the unmanned system. To address these issues, some scholars have suggested prioritizing the experience playback pool and using transfer learning to apply previously learned strategies to new environments. However, these methods only alleviate the speed at which the unmanned system forgets its experiences and do not fundamentally solve the problem. Additionally, they cannot prevent dangerous actions and falling into local optima. Therefore, we propose a dual decision-making continuous learning method based on simulation to reality (Sim2Real). This method employs a knowledge body to eliminate the local optimal dilemma, and corrects bad strategies in a timely manner to ensure that the unmanned system makes the best decision every time. Our experimental results demonstrate that our method has a 30% higher success rate than other state-of-the-art methods, and the model transfer to real scenes is still highly effective. [ABSTRACT FROM AUTHOR]

Combating misinformation is one of the urgent social crises. Much research has shown that disinformation can lead to social panic and adversely affect society. It is crucial to promptly detect and counteract misinformation to reduce its adverse effects. Although progress in text-based fact verification has been made, the community needs further exploration into the user-oriented results. To address this gap, we integrate theories from linguistics, journalism, psychology, and cognitive science. We propose a disinformation detection algorithm based on multidimensional content analysis. This algorithm combines human factors and user perception in text interactive media to establish six dimensions for comprehensive content analysis. We have proposed a quantitative calculation method corresponding to six dimensions to detect misinformation. The average accuracy of the proposed model test on four datasets is 95.28%. The results show that this algorithm can effectively analyze from a multidisciplinary theoretical perspective and effectively identify misinformation in Chinese and English.

Combating misinformation is one of the urgent social crises. Much research has shown that disinformation can lead to social panic and adversely affect society. It is crucial to promptly detect and counteract misinformation to reduce its adverse effects. Although progress in text-based fact verification has been made, the community needs further exploration into the user-oriented results. To address this gap, we integrate theories from linguistics, journalism, psychology, and cognitive science. We propose a disinformation detection algorithm based on multidimensional content analysis. This algorithm combines human factors and user perception in text interactive media to establish six dimensions for comprehensive content analysis. We have proposed a quantitative calculation method corresponding to six dimensions to detect misinformation. The average accuracy of the proposed model test on four datasets is 95.28%. The results show that this algorithm can effectively analyze from a multidisciplinary theoretical perspective and effectively identify misinformation in Chinese and English. [ABSTRACT FROM AUTHOR]

Industry 4.0 focuses on the realization of smart manufacturing based on cyber-physical systems (CPS). However, emerging Industry 5.0 and Society 5.0 reaches beyond CPS and covers the entire value chain of manufacturing, and faces economic, environmental, and social challenges. To meet such challenges, we regard Industry 5.0 as a socio-technical revolution based on the socio-cyber-physical system (SCPS), and propose a socio-technically enhanced wisdom manufacturing architecture and framework beyond CPS-based Industry 4.0/smart manufacturing with especially concerning transition enabling technologies such as artificial intelligence, social Internet of Things (SIoT), big data, machine learning, edge computing, social computing, 3D printing, blockchains, digital twins, and cobots. Finally we address the roadmap to blockchainized value-added SCPS-based Industrial Metaverse for Industry/Society 5.0, which will achieve high utilization of resources and provide products and services to satisfy experience-driven individual needs via metamanufacturing cloud services towards smart, resilient, sustainable, and human-centric solutions. [ABSTRACT FROM AUTHOR]

As consumer adoption of immersive technologies grows, virtual avatars will play a prominent role in the future of social computing. However, as people begin to interact more frequently through virtual avatars, it is important to ensure that the research community has validated tools to evaluate the effects and consequences of such technologies. We present the first iteration of a new, freely available 3D avatar library called the Virtual Avatar Library for Inclusion and Diversity (VALID), which includes 210 fully rigged avatars with a focus on advancing racial diversity and inclusion. We also provide a detailed process for creating, iterating, and validating avatars of diversity. Through a large online study (n = 132) with participants from 33 countries, we provide statistically validated labels for each avatar's perceived race and gender. Through our validation study, we also advance knowledge pertaining to the perception of an avatar's race. In particular, we found that avatars of some races were more accurately identified by participants of the same race. [ABSTRACT FROM AUTHOR]

Abstract Smart grids are typically modelled as cyber–physical power systems, with limited consideration given to the social aspects. Specifically, traditional power system studies tend to overlook the behaviour of stakeholders, such as end‐users. However, the impact of end‐users and their behaviour on power system operation and response to disturbances is significant, particularly with respect to demand response and distributed energy resources. Therefore, it is essential to plan and operate smart grids by taking into account both the technical and social aspects, given the crucial role of active and passive end‐users, as well as the intermittency of renewable energy sources. In order to optimize system efficiency, reliability, and resilience, it is important to consider the level of cooperation, flexibility, and other social features of various stakeholders, including consumers, prosumers, and microgrids. This article aims to address the gaps and challenges associated with modelling social behaviour in power systems, as well as the human‐centred approach for future development and validation of socio‐technical power system models. As the cyber–physical–social system of energy emerges as an important topic, it is imperative to adopt a human‐centred approach in this domain. Considering the significance of computational social science for power system applications, this article proposes a list of research topics that must be addressed to improve the reliability and resilience of power systems in terms of both operation and planning. Solving these problems could have far‐reaching implications for power systems, energy markets, community usage, and energy strategies.

Given that Chinese society is currently undergoing a comprehensive transitional phase characterized by intensifying stratification and conflicting interests among social classes, which has led to the increase of self-initiated events aimed at interest maintenance. The ensuing collective behaviors consequently pose considerable challenges to maintaining social order and stability in China. With the aim of adequately mitigating and forestalling potentially deleterious collective behavior. It is of great significance to study the evolutionary logic of collective behaviors under different circumstances, and thus provide a scientific reference for effectively controlling and preventing the malignant events of collective behaviors. The evolution of collective behaviors is a complex public-administration problem with nonlinear characteristics. Domestic scholars have put forward a series of explanatory frameworks for grassroots populace involvement in collective behaviors by integrating western theories with Chinese context.However, traditional methodologies of social science research can hardly understand the emergence characteristics and the overall structure of collective behaviors from the perspective of microcosmic units, and it is difficult to accurately analyze the specific social mechanism. Based on the existing theories and research of collective behaviors,and leveraging computational sociological thinking and approaches in an innovative manner,we firstly constructed a comprehensive interpretation framework generalizing the evolution of collective behavior into different dimensions of the rational-irrational and independent-influenced ones, and reconstructs the evolutionary logic of collective behaviors in theory combining with the discussion of information receiving and processing mechanism. Four kinds of the information diffusion mechanisms including the global-complete, local-complete, global-incomplete and local-incomplete evolutions were then formed. The complex network theory and information diffusion models in the field of social computing were further introduced, and the process of the evolved behaviors with the modeling and mathematical expression was built, and then a strong operational overall framework for systematically analyzing collective behaviors was formed. Secondly, under the guidance of the overall framework, the analysis framework and evolution models under different dimensions and diffusion mechanisms were constructed, and their specific process was also analyzed mathematically. The result shows that significant social influence and time limitation were identified as environmental constraints that prevent such self-correction and lead to unpredictability and path dependence of the system. In structured crowds, sparse networks facilitate the dissipation of collective behavior, while scale-free networks enhance the self-reinforcement of social influence. Based on the simulation of computer-generated parameters and the investigation data of migrant workers, the mechanisms of both generalization and specialization of collective behaviors were discussed. The scale-free network structure and formal network structure of migrant workers manifest similar characteristics in collective behavior evolution. Migrant workers at critical positions in the formal network have major impacts on the initiation of collective behavior.The research provides targeted perspectives for government agencies to develop contingency plans for collective behaviors as well as construct a modernized governance system led by the government with multi-stakeholder participation from enterprises, social organizations and citizens. [ABSTRACT FROM AUTHOR]

Stance detection has garnered considerable attention from researchers due to its broad range of applications, including fact-checking and social computing. While state-of-the-art stance detection models are usually based on supervised machine learning methods, their effectiveness is heavily reliant on the quality of training data. This problem is more prevalent in stance detection task because the stance of a text is intimately tied to the target under consideration. While numerous datasets exist for stance detection, determining their suitability for a specific target can be challenging. In this work, we focus on Turkish stance detection and explore the impact of training data on the model performance. In particular, we fine-tune BERT model with various datasets and assess their performance when the test data is the same/different compared to the training data in terms of target and domain. In addition, given the scarcity of resources for Turkish stance detection, we investigate i) whether we can use existing datasets in other languages in a cross-lingual setup, and ii) the effectiveness of data augmentation with simple automatic labeling methods. In order to conduct our experiments, we also create new Turkish stance detection datasets for various targets in different domains. In our comprehensive experiments, our findings are as follows. 1) Using training data with multiple targets in the same domain yields high performance as the model is able to learn more characteristics of expressing stance with additional data. 2) The domain of the training data plays a crucial role in achieving high performance. 3) Automatically generated data enhances performance when combined with manually annotated data. 4) Training solely on Turkish data outperforms training with the combination of Turkish and English data. Overall, our study points out the importance of creating Turkish annotated datasets for different domains to achieve high performance in stance detection. [ABSTRACT FROM AUTHOR]

Influence Maximization (IM) is a classical combinatorial optimization problem, which can be widely used in mobile networks, social computing, and recommendation systems. It aims at selecting a small number of users such that maximizing the influence spread across the online social network. Because of its potential commercial and academic value, there are a lot of researchers focusing on studying the IM problem from different perspectives. The main challenge comes from the NP-hardness of the IM problem and #P-hardness of estimating the influence spread, thus traditional algorithms for overcoming them can be categorized into two classes: heuristic algorithms and approximation algorithms. However, there is no theoretical guarantee for heuristic algorithms, and the theoretical design is close to the limit. Therefore, it is almost impossible to further optimize and improve their performance. With the rapid development of artificial intelligence, technologies based on Machine Learning (ML) have achieved remarkable achievements in many fields. In view of this, in recent years, a number of new methods have emerged to solve combinatorial optimization problems by using ML-based techniques. These methods have the advantages of fast solving speed and strong generalization ability to unknown graphs, which provide a brand-new direction for solving combinatorial optimization problems. Therefore, we abandon the traditional algorithms based on iterative search and review the recent development of ML-based methods, especially Deep Reinforcement Learning, to solve the IM problem and other variants in social networks. We focus on summarizing the relevant background knowledge, basic principles, common methods, and applied research. Finally, the challenges that need to be solved urgently in future IM research are pointed out. [ABSTRACT FROM AUTHOR]

Personality-aware recommendation systems have been proven to achieve high accuracy compared to conventional recommendation systems. In addition to that, personality-aware recommendation systems could help alleviate cold start and data sparsity problems by adding the user's personality traits in the recommendation process. The majority of the literature works used Big-Five personality model to represent the user's personality, this is due to the popularity of Big-Five model in the literature of psychology. However, from personality computing perspective, the choice of the most suitable personality model that satisfy the requirements of the recommendation application and the recommended content type still needs further investigation. In this paper, we study and compare four personality-aware recommendation systems based on different personality models, namely Big-Five traits model, Eysenck model and HEXACO model from the personality traits theory, and Myers–Briggs Type Indicator (MPTI) from the personality types theory. Furthermore, we propose a hybrid personality model for recommendation that takes advantage of the personality traits models, as well as the personality types models. Through extensive experiments on recommendation dataset, we prove the efficiency of the proposed model, especially in cold start settings. Our proposed hybrid personality-aware recommendation model improves the precision and recall in cold start settings by 21% and 18% respectively compared to the widely used Big-Five traits model. [ABSTRACT FROM AUTHOR]

The fairness of human‐related software has become critical with its widespread use in our daily lives, where life‐changing decisions are made. However, with the use of these systems, many erroneous results emerged. Technologies have started to be developed to tackle unexpected results. As for the solution to the issue, companies generally focus on algorithm‐oriented errors. The utilized solutions usually only work in some algorithms. Because the cause of the problem is not just the algorithm; it is also the data itself. For instance, deep learning cannot establish the cause–effect relationship quickly. In addition, the boundaries between statistical or heuristic algorithms are unclear. The algorithm's fairness may vary depending on the data related to context. From this point of view, our article focuses on how the data should be, which is not a matter of statistics. In this direction, the picture in question has been revealed through a scenario specific to "vulnerable and disadvantaged" groups, which is one of the most fundamental problems today. With the joint contribution of computer science and social sciences, it aims to predict the possible social dangers that may arise from artificial intelligence algorithms using the clues obtained in this study. To highlight the potential social and mass problems caused by data, Gerbner's "cultivation theory" is reinterpreted. To this end, we conduct an experimental evaluation on popular algorithms and their data sets, such as Word2Vec, GloVe, and ELMO. The article stresses the importance of a holistic approach combining the algorithm, data, and an interdisciplinary assessment. This article is categorized under:Algorithmic Development > Statistics [ABSTRACT FROM AUTHOR]

The following review incorporates a systematic selection, coding, and analysis methodology in order to compile a corpus of empirical research studies that investigate the use of technology-mediated feedback in L2 writing contexts published from 2015-2019. Trends are identified by coding and quantitatively analyzing key parameters of the corpus, and the corpus is qualitatively analyzed, contextualizing and exploring emergent themes within these trends in rich detail. Key quantitative findings reveal (A) the leading journals are Computer Assisted Language Learning, Computers & Education, Journal of Second Language Writing, and Language Learning & Technology; (B) the region with the most publications is Greater Asia, with the USA as the country with the most publications; (C) the majority of studies were conducted in EFL contexts; (D) tertiary contexts dominate, with the majority occurring at the undergraduate level in blended learning contexts; (E) the majority of studies examined feedback that contained both local and global feedback; (F) the majority of studies gave comprehensive feedback via a mix of direct and indirect feedback, were delivered within formative treatments, and used either asynchronous or mixtures of both asynchronous and synchronous modes; (G) as for types of technology used, multiple technologies were used most, followed by individual study tools, cloud-based word processors and shared documents, and network-based social computing; (H) most research orientations were descriptive or comparative and used non-experimental designs; and (I) the most popular sources of feedback were issued from teachers, AWEs, and peers. Eight studies investigated the use of feedback stemming from multiple sources. Finally, qualitative analyses explore emergent themes involving the intersection of source of feedback and technology type; in doing so, the authors describe, contextualize, and critique in rich detail how the use of artful combinations of technologies and sources are currently shaping the use of technology-mediated feedback in L2 writing instruction and research. [ABSTRACT FROM AUTHOR]

The automatic evaluation of the teaching effect has been a technical problem for many years. Because only video frames are available for it, and the information extraction from such dynamic scenes still remains challenging. In recent years, the progress of deep learning has boosted the application of computer vision in many areas, which can provide much insight into the above issue. As a consequence, this paper proposes a vision sensing-based automatic evaluation method for teaching effects based on deep residual network (DRN). The DRN is utilized to construct a backbone network for sensing from visual features such as attending status, taking notes, playing phones, looking outside, etc. The extracted visual features are further selected as the basis for the evaluation of the teaching effect. We have also collected some realistic course images to establish a real-world dataset for the performance assessment of the proposal. The proposed method is implemented on collected datasets via computer programming-based simulation experiments, so as to obtain accuracy assessment results as measurement. The obtained results show that the proposal can well perceive typical visual features from video frames of courses and realize automatic evaluation of the teaching effect.

The COVID-19 pandemic has severely harmed every aspect of our daily lives, resulting in a slew of social problems. Therefore, it is critical to accurately assess the current state of community functionality and resilience under this pandemic for successful recovery. To this end, various types of social sensing tools, such as tweeting and publicly released news, have been employed to understand individuals’ and communities’ thoughts, behaviors, and attitudes during the COVID-19 pandemic. However, some portions of the released news are fake and can easily mislead the community to respond improperly to disasters like COVID-19. This paper aims to assess the correlation between various news and tweets collected during the COVID-19 pandemic on community functionality and resilience. We use fact-checking organizations to classify news as real, mixed, or fake, and machine learning algorithms to classify tweets as real or fake to measure and compare community resilience (CR). Based on the news articles and tweets collected, we quantify CR based on two key factors, community wellbeing and resource distribution, where resource distribution is assessed by the level of economic resilience and community capital. Based on the estimates of these two factors, we quantify CR from both news articles and tweets and analyze the extent to which CR measured from the news articles can reflect the actual state of CR measured from tweets. To improve the operationalization and sociological significance of this work, we use dimension reduction techniques to integrate the dimensions.

During the past decades, the term “social computing” has become a promising interdisciplinary area in the intersection of computer science and social science. In this work, we conduct a data-driven study to understand the development of social computing using the data collected from Digital Bibliography and Library Project (DBLP), a representative computer science bibliography website. We have observed a series of trends in the development of social computing, including the evolution of the number of publications, popular keywords, top venues, international collaborations, and research topics. Our findings will be helpful for researchers and practitioners working in relevant fields.