Search

Nowadays, adolescents would like to share their daily lives via social media platforms, which presents an excellent opportunity for us to leverage these data to develop techniques to measure their mental health status, such as depression. Previous researches focus on the more accurate detection of depression through statistical learning and ignore psychological understanding of depression. However, psychologists have given lots of theoretical evidence for depression. Such as according to cognitive psychology research, cognitive distortions will result in depression. Thus, in this study, we propose a new task, explainable depression detection, to not only automatically detect depression but also try to give clues to depression based on cognitive distortion theory. For this purpose, we construct a multi-task learning model based on a pre-trained model to detect depression and identify cognitive distortion. And we use many analytical means including word clouds for data analysis to draw our conclusion. Previous social media users' depression corpus and our cognitive distortion corpus are utilized for analysis and experiment. Our experimental results outperform the baseline results and interesting conclusions about adolescent depression are drawn., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2023 Wang, Zhao, Lu and Qin.)

Autonomous vehicles (AVs) will bring considerable benefits to individuals and society, while the process of AVs' popularity may not always be smooth. Sometimes crashes are unavoidable, which could hinder the widespread acceptance of AVs. Although previous studies have investigated factors that affect AV acceptance, few of them pay attention to the impact of accidents and not to mention quantitative analysis. This study aims to explore what the public is really concerned about the AV crash and whether it significantly affects the public's behavioral intention to use AVs utilizing a multi-method analytical approach, including latent Dirichlet allocation, structural equation model, and Bayesian network. We employed a qualitative exploratory analysis to identify the topics of social media data about AV crashes and proposed an integrated framework using variables originating from topics. We introduced perceived severity in AV acceptance research to cater to the context of AV crashes. The findings reveal how trust, attitude, knowledge, perceived risk, media exposure, and perceived severity affect the public acceptance of AVs and how differences exist between full automation and driver assistance. This research helps to provide theoretical contributions and practical implications to reduce the negative effect of AV crashes and improve public acceptance of AVs. • This paper investigates public perceptions of autonomous vehicles (AVs) after an accident. • A hybrid method combining social media and a questionnaire survey aims to get the real public concern about AV crashes. • We proposed an integrated framework using variables originating from social media data. • This study reveals the mechanism of consumers' adoption of AV technology. [ABSTRACT FROM AUTHOR]

• Understand public perception of AVs by analyzing social media data in China; • Introduce text network analysis to the research on public perception of AVs as a supplement to existing research methods; • Analyze the gender differences in public perception of AVs using social media; • The public's insufficient knowledge of AVs is related to the publicity of automobile companies and the media; • The public expresses more blame for vehicles with L2 driving automation than human drivers after the NIO crash. In China, vehicles with L2 driving automation are already widespread, while L3-L5 automated vehicles (AVs) are not allowed on the road. Surprisingly, crashes involving vehicles with L2 driving automation contributed to the public's discussion of L3-L5 AVs. Understanding the variations in public perception of AVs is imperative, which could significantly affect the future evolution of AVs. Respondents have few opportunities to express their views freely in survey-based studies relying on structured questionnaires (e.g., Likert scale). We collected 42,111 comments from Chinese mainstream social media platforms (Sina Weibo and Tik Tok) in August 2021 and applied advanced text mining technology (Latent Dirichlet Allocation topic model, text network analysis, and sentiment analysis) to understand variations in public perception of AVs before and after the crashes. The results show that the public expresses more blame for vehicles with L2 driving automation than human drivers and has misunderstandings about the current development of L3-L5 AVs, which is closely connected with publicity from automobile companies and media. Women were more negative than men both before and after the crash, and the sentiment difference was most significant in the first three days of the crash. This research provides new ideas for a timely understanding of the public perception of AVs. Findings contribute to specific policy recommendations to help in the technological change and wider adoption of AVs. [ABSTRACT FROM AUTHOR]