Search results

The aim of this study is to investigate subject independent emotion recognition capabilities of EEG and peripheral physiological signals namely: electroocoulogram (EOG), electromyography (EMG), electrodermal activity (EDA), temperature, plethysmograph and respiration. The experiments are conducted on both modalities independently and in combination. This study arranges the physiological signals in order based on the prediction accuracy obtained on test data using time and frequency domain features.

DEAP dataset is used in this experiment. Time and frequency domain features of EEG and physiological signals are extracted, followed by correlation-based feature selection. Classifiers namely – Naïve Bayes, logistic regression, linear discriminant analysis, quadratic discriminant analysis, logit boost and stacking are trained on the selected features. Based on the performance of the classifiers on the test set, the best modality for each dimension of emotion is identified.

 The experimental results with EEG as one modality and all physiological signals as another modality indicate that EEG signals are better at arousal prediction compared to physiological signals by 7.18%, while physiological signals are better at valence prediction compared to EEG signals by 3.51%. The valence prediction accuracy of EOG is superior to zygomaticus electromyography (zEMG) and EDA by 1.75% at the cost of higher number of electrodes. This paper concludes that valence can be measured from the eyes (EOG) while arousal can be measured from the changes in blood volume (plethysmograph). The sorted order of physiological signals based on arousal prediction accuracy is plethysmograph, EOG (hEOG + vEOG), vEOG, hEOG, zEMG, tEMG, temperature, EMG (tEMG + zEMG), respiration, EDA, while based on valence prediction accuracy the sorted order is EOG (hEOG + vEOG), EDA, zEMG, hEOG, respiration, tEMG, vEOG, EMG (tEMG + zEMG), temperature and plethysmograph.

Many of the emotion recognition studies in literature are subject dependent and the limited subject independent emotion recognition studies in the literature report an average of leave one subject out (LOSO) validation result as accuracy. The work reported in this paper sets the baseline for subject independent emotion recognition using DEAP dataset by clearly specifying the subjects used in training and test set. In addition, this work specifies the cut-off score used to classify the scale as low or high in arousal and valence dimensions. Generally, statistical features are used for emotion recognition using physiological signals as a modality, whereas in this work, time and frequency domain features of physiological signals and EEG are used. This paper concludes that valence can be identified from EOG while arousal can be predicted from plethysmograph.

Using a mobile phone is increasingly becoming recognized as very dangerous while driving. With a smartphone, users feel connected and have access to information. The inability to access smartphone has become a phobia, causing anxiety and fear. The present study’s aims are as follows: first, quantify the association between nomophobia and road safety among motorists; second, determine a cut-off value for nomophobia that would identify poor road safety so that interventions can be designed accordingly.

Participants were surveyed online for nomophobia symptoms and a recent history of traffic contraventions. Nomophobia was measured using the nomophobia questionnaire (NMP-Q).

A total of 1731 participants responded to the survey; the mean age was 33 ± 12, and 43% were male. Overall, 483 (28%) [26–30%] participants received a recent traffic contravention. Participants with severe nomophobia showed a statistically significant increased risk for poor road safety odds ratios and a corresponding 95% CI of 4.64 [3.35-6.38] and 4.54 [3.28-6.29] in crude and adjusted models, respectively. Receiver operator characteristic (ROC)-based analyses revealed that NMP-Q scores of = 90 would be effective for identifying at risk drivers with sensitivity, specificity and accuracy of 61%, 75% and 72%, respectively.

Nomophobia symptoms are quite common among adults. Severe nomophobia is associated with poor road safety among motorists. Developing screening and intervention programs aimed at reducing nomophobia may improve road safety among motorists.

The purpose of this paper is to investigate the comment letters (CLs) in the standard-setting process of audits of less complex entities (LCEs). The objective is to gain insight into the overall picture of the CLs and to report on areas where comment providers agree or disagree with IAASB's Part 10.

A content analysis of 60 comment letter (CLs) was conducted to investigate the suggested additional Part 10 on audits of groups' financial statements in the proposed ISA for LCEs. Hence, this study examines three specific topics: (1) the views related to the use of the International Standard on Auditing (ISA) for LCEs for group audits in which component auditors are involved, (2) the proposed group-specific qualitative characteristics to describe the scope of group audits and, finally, (3) insights into the content of the proposed Part 10 and related conforming amendments. The Gioia method is used to provide a holistic approach to concept development of the arguments about the new Part 10.

The CLs stated that, while the proposed Part 10 has some weak points, it still provides a solid and practical structure within which to undertake an LCE group audit and a promising basis for further development. For instance, when discussing the improvements, the CLs stated that Part 10 should allow for more auditor judgment when determining when the involvement of component auditors renders a group audit complex. In addition, the CLs asserted that professional judgment should be engaged when considering the qualitative characteristics and the complexity of the group.

This study contributes to the very scarce research about the ISA for LCEs and the role of lobbying in shaping the audit standard-setting process.

Considering the ongoing discourse on diversity, equity and inclusion, brands aim to develop marketing campaigns that demonstrate respect for all individuals. Despite these intentions, many advertisements still provoke strong negative reactions from consumers due to brand transgressions in social media marketing campaigns that violate these values. The purpose of this paper is to analyze the repercussions that such social media marketing campaigns have on brands, categorizing these campaigns as brand transgressions in social media advertising.

This research uses a mixed-method design that includes semi-structured interviews (Study 1), a content analysis (Study 2) and an online experiment (Study 3).

This paper clarifies the elements that qualify as brand transgressions in advertising within the diversity, equity and inclusion discourse. The negative electronic word-of-mouth (e-WOM) associated with brand transgressions in advertising comprises negative emotions (e.g. anger, contempt, disgust and hate) and behavioural intentions to penalize the brand (e.g. negative word-of-mouth, brand avoidance and protest behaviours). The negative e-WOM stemming from these transgressions amplifies the adverse consequences for consumer–brand relationships by negatively influencing other consumers through sympathy towards the offended parties.

This paper offers brand managers guidelines for preventing and managing negative consumer reactions towards brands based on their responses to marketing campaigns that contradict the principles of diversity, equity and inclusion.

This paper contributes to the literature on brand transgressions related to diversity, equity and inclusion values by exploring their impact on consumer–brand relationships and highlighting the pivotal role of sympathy in perpetuating negative consequences.

Numerous design methods are available to facilitate digital innovation processes in user interface design. Nonetheless, little guidance exists on their appropriate selection within the design process based on specific situations. Consequently, design novices with limited design knowledge face challenges when determining suitable methods. Thus, this paper aims to support design novices by guiding the situational selection of design methods.

Our research approach includes two phases: i) we adopted a taxonomy development method to identify dimensions of design methods by reviewing 292 potential design methods and interviewing 15 experts; ii) we conducted focus groups with 25 design novices and applied fuzzy-set qualitative comparative analysis to describe the relations between the taxonomy's dimensions.

We developed a novel taxonomy that presents a comprehensive overview of design conditions and their associated design methods in innovation processes. Thus, the taxonomy enables design novices to navigate the complexities of design methods needed to design digital innovation. We also identify configurations of these conditions that support the situational selections of design methods in digital innovation processes of user interface design.

The study’s contribution to the literature lies in the identification of both similarities and differences among design methods, as well as the investigation of sufficient condition configurations within the digital innovation processes of user interface design. The taxonomy helps design novices to navigate the design space by providing an overview of design conditions and the associations between methods and these conditions. By using the developed taxonomy, design novices can narrow down their options when selecting design methods for their specific situations.

Automatically extracting stance information from natural language texts is a significant research problem with various applications, particularly after the recent explosion of data on the internet via platforms like social media sites. Stance detection system helps determine whether the author agree, against or has a neutral opinion with the given target. Most of the research in stance detection focuses on the English language, while few research was conducted on the Arabic language.

This paper aimed to address stance detection on Arabic tweets by building and comparing different stance detection models using four transformers, namely: Araelectra, MARBERT, AraBERT and Qarib. Using different weights for these transformers, the authors performed extensive experiments fine-tuning the task of stance detection Arabic tweets with the four different transformers.

The results showed that the AraBERT model learned better than the other three models with a 70% F1 score followed by the Qarib model with a 68% F1 score.

A limitation of this study is the imbalanced dataset and the limited availability of annotated datasets of SD in Arabic.

Provide comprehensive overview of the current resources for stance detection in the literature, including datasets and machine learning methods used. Therefore, the authors examined the models to analyze and comprehend the obtained findings in order to make recommendations for the best performance models for the stance detection task.

Sports tourism was strongly affected by the COVID-19 pandemic, but there is no consensus on what sports tourism should look like in the post-pandemic period. This study explores the future of sports tourism in light of the COVID-19 pandemic and provides an alternative paradigm model.

Data were collected by interviewing sports tourism experts. Data analysis was based on the continuous comparison method during three stages of open, axial and selective coding.

The findings point to the complexity of the future sports tourism industry. Post-COVID-19 sports tourism strongly depends on environmental forces and targeted support, with strategies focused on tourists’ safety and security, digitalization of the industry, and new employment opportunities.

This study contributes to the body of knowledge on sports tourism by providing answers to the current challenges, threats and opportunities associated with the pandemic. The proposed paradigm model could be a guideline for sports tourism practitioners and policymakers to accelerate recovery from COVID-19 in a sustainable and resilient manner.

Social networks (SNs) have recently evolved from a means of connecting people to becoming a tool for social engineering, radicalization, dissemination of propaganda and recruitment of terrorists. It is no secret that the majority of the Islamic State in Iraq and Syria (ISIS) members are Arabic speakers, and even the non-Arabs adopt Arabic nicknames. However, the majority of the literature researching the subject deals with non-Arabic languages. Moreover, the features involved in identifying radical Islamic content are shallow and the search or classification terms are common in daily chatter among people of the region. The authors aim at distinguishing normal conversation, influenced by the role religion plays in daily life, from terror-related content.

This article presents the authors' experience and the results of collecting, analyzing and classifying Twitter data from affiliated members of ISIS, as well as sympathizers. The authors used artificial intelligence (AI) and machine learning classification algorithms to categorize the tweets, as terror-related, generic religious, and unrelated.

The authors report the classification accuracy of the K-nearest neighbor (KNN), Bernoulli Naive Bayes (BNN) and support vector machine (SVM) [one-against-all (OAA) and all-against-all (AAA)] algorithms. The authors achieved a high classification F1 score of 83\%. The work in this paper will hopefully aid more accurate classification of radical content.

In this paper, the authors have collected and analyzed thousands of tweets advocating and promoting ISIS. The authors have identified many common markers and keywords characteristic of ISIS rhetoric. Moreover, the authors have applied text processing and AI machine learning techniques to classify the tweets into one of three categories: terror-related, non-terror political chatter and news and unrelated data-polluting tweets.

This paper analyses the factors that make fashion-product crowdfunding campaigns successful. The authors argue that crowdfunding is an innovative and functional way of bringing new fashion items to the market. The purpose of this paper is to answer the question whether product innovation, lifecycle and sustainability have a positive effect on the success of fashion crowdfunding campaigns. The findings highlight that the success of the fashion crowdfunding campaigns depends on creators' adherence to the values of the platform which they use to raise capital.

A total of 300 fashion crowdfunding projects running between the 17th of October and the 15th of December 2017 were collected from Kickstarter – the world's largest crowdfunding platform based on reward-based all-or-nothing model. Two-step binomial logistic regression was used to analyse the data.

The model predicted a significant increase in the odds of success for the fashion items crowdfunded during the first-time production, and innovative and environmentally sustainable products with a higher price range of rewards. In line with previous literature, regression analyses predicted a significant effect of the control variables of goal amount (negative) and the number of rewards (positive). Contrary to previous studies, neither the presence of a video nor the campaign length predicted success.

The novel findings of this study contribute to the literature by providing an analysis of success factors of fashion items on crowdfunding platforms. The results show that innovative, environmentally sustainable and higher-priced products produced by early-stage ventures are better welcomed by the audiences.

This study is motivated to provide a theoretically informed, data-driven assessment of the consequences associated with the participation of non-human bots in social accountability movements; specifically, the anti-inequality/anti-corporate #OccupyWallStreet conversation stream on Twitter.

A latent Dirichlet allocation (LDA) topic modeling approach as well as XGBoost machine learning algorithms are applied to a dataset of 9.2 million #OccupyWallStreet tweets in order to analyze not only how the speech patterns of bots differ from other participants but also how bot participation impacts the trajectory of the aggregate social accountability conversation stream. The authors consider two research questions: (1) do bots speak differently than non-bots and (2) does bot participation influence the conversation stream.

The results indicate that bots do speak differently than non-bots and that bots exert both weak form and strong form influence. Bots also steadily become more prevalent. At the same time, the results show that bots also learn from and adapt their speaking patterns to emphasize the topics that are important to non-bots and that non-bots continue to speak about their initial topics.

These findings help improve understanding of the consequences of bot participation within social media-based democratic dialogic processes. The analyses also raise important questions about the increasing importance of apparently nonhuman actors within different spheres of social life.

The current study is the first, to the authors’ knowledge, that uses a theoretically informed Big Data approach to simultaneously consider the micro details and aggregate consequences of bot participation within social media-based dialogic social accountability processes.