Search

Social media represent an important source of news for many users. They are, however, affected by misinformation and they might be playing a role in the growth of political polarization. In this paper, we create an agent based model to investigate how policing content and backlash on social media (i.e. conflict) can lead to an increase in polarization for both users and news sources. Our model is an advancement over previously proposed models because it allows us to study the polarization of both users and news sources, the evolution of the audience connections between users and sources, and it makes more realistic assumptions about the starting conditions of the system. We find that the tendency of users and sources to avoid policing, backlash and conflict in general can increase polarization online. Specifically polarization comes from the ease of sharing political posts, intolerance for opposing points of view causing backlash and policing, and volatility in changing one’s opinion when faced with new information. On the other hand, it seems that the integrity of a news source in trying to resist the backlash and policing has little effect.

Emotion dynamics is a framework for measuring how an individual's emotions change over time. It is a powerful tool for understanding how we behave and interact with the world. In this paper, we introduce a framework to track emotion dynamics through one's utterances. Specifically we introduce a number of utterance emotion dynamics (UED) metrics inspired by work in Psychology. We use this approach to trace emotional arcs of movie characters. We analyze thousands of such character arcs to test hypotheses that inform our broader understanding of stories. Notably, we show that there is a tendency for characters to use increasingly more negative words and become increasingly emotionally discordant with each other until about 90 percent of the narrative length. UED also has applications in behavior studies, social sciences, and public health., 19 pages, 7 figures

Information sharing via social networking systems (SNS) is a common practice among academics, as well as others, that brings substantial benefits. At the same time, privacy concerns are widespread among SNS users, which may tend to inhibit their maximising the benefit from using the systems. This paper investigates the proposition that SNS user attitudes and behaviour are affected by privacy concerns, and that the effects are subject to significant cultural factors. A broad assessment of the literature provides the context for the study. Working in the context of Saudi Arabia, we apply a mixed-methods approach beginning with in-depth interviews, exposing in detail a range of views and concerns about privacy and SNS use, also allowing us to identify three key factors that bear on SNS usage and users’ concerns. Analysis of these factors in the light of the “theory of reasoned action” derives a structural model predicting several hypotheses relating the factors and users’ attitudes and behaviour. We assess the model through development of a questionnaire, administered to a large pool of academic participants, that allows us to examine how the model responds in general, and via multigroup partial least squares analyses, differentially to gender and to culturally distinct (Arab vs. non-Arab) constituents of the participant group. Results show good support for the hypotheses and clear gender and culture effects. Picking up issues from the interviews, discussion focuses on users’ views about SNS providers’ privacy policies and their inadequacy regarding culturally specific ethical concerns. We argue that these views may reflect different regulatory environments in combination with other cultural factors.

Despite the prevalence of disagreement between users on social media platforms, studies of online debates typically only look at positive online interactions, represented as networks with positive ties. In this paper, we hypothesize that the systematic neglect of conflict that these network analyses induce leads to misleading results on polarized debates. We introduce an approach to bring in negative user-to-user interaction, by analyzing online debates using signed networks with positive and negative ties. We apply this approach to the Dutch Twitter debate on ‘Black Pete’—an annual Dutch celebration with racist characteristics. Using a dataset of 430,000 tweets, we apply natural language processing and machine learning to identify: (i) users’ stance in the debate; and (ii) whether the interaction between users is positive (supportive) or negative (antagonistic). Comparing the resulting signed network with its unsigned counterpart, the retweet network, we find that traditional unsigned approaches distort debates by conflating conflict with indifference, and that the inclusion of negative ties changes and enriches our understanding of coalitions and division within the debate. Our analysis reveals that some groups are attacking each other, while others rather seem to be located in fragmented Twitter spaces. Our approach identifies new network positions of individuals that correspond to roles in the debate, such as leaders and scapegoats. These findings show that representing the polarity of user interactions as signs of ties in networks substantively changes the conclusions drawn from polarized social media activity, which has important implications for various fields studying online debates using network analysis.

The dimensionality of the spatially distributed channels and the temporal resolution of electroencephalogram (EEG) based brain-computer interfaces (BCI) undermine emotion recognition models. Thus, prior to modeling such data, as the final stage of the learning pipeline, adequate preprocessing, transforming, and extracting temporal (i.e., time-series signals) and spatial (i.e., electrode channels) features are essential phases to recognize underlying human emotions. Conventionally, inter-subject variations are dealt with by avoiding the sources of variation (e.g., outliers) or turning the problem into a subject-deponent. We address this issue by preserving and learning from individual particularities in response to affective stimuli. This paper investigates and proposes a subject-independent emotion recognition framework that mitigates the subject-to-subject variability in such systems. Using an unsupervised feature selection algorithm, we reduce the feature space that is extracted from time-series signals. For the spatial features, we propose a subject-specific unsupervised learning algorithm that learns from inter-channel co-activation online. We tested this framework on real EEG benchmarks, namely DEAP, MAHNOB-HCI, and DREAMER. We train and test the selection outcomes using nested cross-validation and a support vector machine (SVM). We compared our results with the state-of-the-art subject-independent algorithms. Our results show an enhanced performance by accurately classifying human affection (i.e., based on valence and arousal) by 16%–27% compared to other studies. This work not only outperforms other subject-independent studies reported in the literature but also proposes an online analysis solution to affection recognition.

Sentiment analysis is an evolving field of study that employs artificial intelligence techniques to identify the emotions and opinions expressed in a given text. Applying sentiment analysis to study the billions of messages that circulate in popular online social media platforms has raised numerous opportunities for exploring the emotional expressions of their users. In this paper we combine sentiment analysis with natural language processing and topic analysis techniques and conduct two different studies to examine whether engagement in entrepreneurship is associated with more positive emotions expressed on Twitter. In study 1, we investigate three samples with 6.717.308, 13.253.244, and 62.067.509 tweets respectively. We find that entrepreneurs express more positive emotions than non-entrepreneurs for most topics. We also find that social entrepreneurs express more positive emotions, and that serial entrepreneurs express less positive emotions than other entrepreneurs. In study 2, we use 21.491.962 tweets to explore 37.225 job-status changes by individuals who entered or quit entrepreneurship. We find that a job change to entrepreneurship is associated with a shift in the expression of emotions to more positive ones.

Lane detection in complex road scenes is still a challenging task due to poor lighting conditions, interference of irrelevant road markings or signs, etc. To solve the problem of lane detection in the various complex road scenes, we proposed a geometric attention-aware network (GAAN) for lane detection. The proposed GAAN adopted a multi-task branch architecture, and used the attention information propagation (AIP) module to perform communication between branches, then the geometric attention-aware (GAA) module was used to complete feature fusion. In order to verify the lane detection effect of the proposed model in this paper, the experiments were conducted on the CULane dataset, TuSimple dataset, and BDD100K dataset. The experimental results show that our method performs well compared with the current excellent lane line detection networks.

The prevalence of persons living with dementia and at risk of going missing is rising. In this study, we engaged persons living with dementia, care partners, police services, search and rescue organizations, and health and social service providers to develop Community ASAP, a mobile alert system that engages community citizens, as volunteers, to look out for persons with dementia reported missing. We completed three phases of development and evaluation of the usability and functionality of the alert system with stakeholders in three Canadian provinces. In this paper we describe features of the Community ASAP and the findings of these evaluation phases.

Emotions at work have long been identified as critical signals of work motivations, status, and attitudes, and as predictors of various work-related outcomes. When more and more employees work remotely, these emotional signals of workers become harder to observe through daily, face-to-face communications. The use of online platforms to communicate and collaborate at work provides an alternative channel to monitor the emotions of workers. This paper studies how emojis, as non-verbal cues in online communications, can be used for such purposes and how the emotional signals in emoji usage can be used to predict future behavior of workers. In particular, we present how the developers on GitHub use emojis in their work-related activities. We show that developers have diverse patterns of emoji usage, which can be related to their working status including activity levels, types of work, types of communications, time management, and other behavioral patterns. Developers who use emojis in their posts are significantly less likely to dropout from the online work platform. Surprisingly, solely using emoji usage as features, standard machine learning models can predict future dropouts of developers at a satisfactory accuracy. Features related to the general use and the emotions of emojis appear to be important factors, while they do not rule out paths through other purposes of emoji use.

In this paper, a general class of estimators is proposed for estimating the finite population mean for sensitive variable, in the presence of measurement error and non-response in simple random sampling. Expressions for bias and mean square error up to first order of approximation, are derived. Impact of measurement errors is examined using real data sets, including the survey conducted at Quaid-i-Azam University, Islamabad. Simulated data sets are also used to observe the performance of the proposed estimators in comparison to some other estimators. We obtain the empirical bias and MSE values for the proposed and the competing estimators.

Happiness levels often fluctuate from one day to the next, and an exogenous shock such as a pandemic can likely disrupt pre-existing happiness dynamics. This paper fits a Marko Switching Dynamic Regression Model (MSDR) to better understand the dynamic patterns of happiness levels before and during a pandemic. The estimated parameters from the MSDR model include each state’s mean and duration, volatility and transition probabilities. Once these parameters have been estimated, we use the one-step method to predict the unobserved states’ evolution over time. This gives us unique insights into the evolution of happiness. Furthermore, as maximising happiness is a policy priority, we determine the factors that can contribute to the probability of increasing happiness levels. We empirically test these models using New Zealand’s daily happiness data for May 2019 –November 2020. The results show that New Zealand seems to have two regimes, an unhappy and happy regime. In 2019 the happy regime dominated; thus, the probability of being unhappy in the next time period (day) occurred less frequently, whereas the opposite is true for 2020. The higher frequency of time periods with a probability of being unhappy in 2020 mostly correspond to pandemic events. Lastly, we find the factors positively and significantly related to the probability of being happy after lockdown to be jobseeker support payments and international travel. On the other hand, lack of mobility is significantly and negatively related to the probability of being happy.

The paper reports the results of a study into the use of linguistic cues defined as Persuasive Linguistic Tricks (PLT) in social media (SM) marketing communication. It was assumed that the content shared on Social Networking Sites (SNS) could be perceived as specific sets of meanings (memeplexes), where a single component, also PLT, may function as their part. Following an original typology of PLT, created based on an emotional factor, the research focused on whether and how the number of positive, neutral and negative PLT used in Facebook posts impacted the behaviour of content recipients. These activities, including liking, commenting and sharing, are strictly connected with post spreading and range. The data analysis focused on 167 Facebook posts shared by five leading Polish travel agencies and 1911 responding comments. The quantitative content analysis method and Spearman’s correlation tests were used. A relationship was observed between the number of emotionally positive and neutral PLT and the increase in the range of content with these PLT. The use of PLT by post recipients was also observed in their comments. This phenomenon is possibly related to the memetic nature of PLT. From the perspective of marketing messages, the obtained results contribute to and guide the textual content-building with a high spreading potential owing to the memetic capability of PLT. Further elaborations were made on the assumption for the evolutionary approach in social media content transfer and its processing.