Motivations for user satisfaction of mobile fitness applications: An analysis of user experience based on online review comments.

Considering that mobile fitness applications are one of the necessities in our lives, the user perspective toward the application is a prominent research topic in both academia and industry with the goal of improving such services. Thus, this study applies two different natural language processing approaches, bag-of-words, and sentiment analysis, to online review comments of the applications to examine the effects of user experience elements. The review dataset collected from 16,461 users, after pre-processing, revealed the notable roles of perceived affection and hedonic values in determining user satisfaction with the application, whereas the effect of user burden on satisfaction was marginal. Several implications, as well as limitations of the study, were examined incorporating the findings with the statistical results.

Computational analysis of user experience and customer satisfaction with mobile food delivery services: Evidence from big data approaches.

Because of the COVID-19 global pandemic, mobile food delivery services have gained new prominence in our society. With this trend, the understanding of user experience in improving mobile food delivery services has gained increasing importance. To this end, we explore how user experience factors extracted by two natural language processing methods from comments of user reviews of mobile food delivery services significantly improve user satisfaction with the services. The results of two multiple regression analyses show that sentiment dimension factors, as well as usability, usefulness, and affection, have notable effects on satisfaction with the applications. Based on several findings of this study, we examine the significant implications and present the limitations of the study.

FibVID: Comprehensive fake news diffusion dataset during the COVID-19 period.

As the SARS-CoV-2 (COVID-19) pandemic has run rampant worldwide, the dissemination of misinformation has sown confusion on a global scale. Thus, understanding the propagation of fake news and implementing countermeasures has become exceedingly important to the well-being of society. To assist this cause, we produce a valuable dataset called FibVID (Fake news information-broadcasting dataset of COVID-19), which addresses COVID-19 and non-COVID news from three key angles. First, we provide truth and falsehood (T/F) indicators of news items, as labeled and validated by several fact-checking platforms (e.g., Snopes and Politifact). Second, we collect spurious-claim-related tweets and retweets from Twitter, one of the world's largest social networks. Third, we provide basic user information, including the terms and characteristics of "heavy fake news" user to present a better understanding of T/F claims in consideration of COVID-19. FibVID provides several significant contributions. It helps to uncover propagation patterns of news items and themes related to identifying their authenticity. It further helps catalog and identify the traits of users who engage in fake news diffusion. We also provide suggestions for future applications of FibVID with a few exploratory analyses to examine the effectiveness of the approaches used.