In today's digital age, Massive Open Online Courses (MOOCs) break the time and space constraints of face-to-face teaching and are being widely used in higher education (Daradoumis et al., 2013). Many top universities participate in and offer high-quality online courses on MOOC learning platforms such as Coursera and edX. It is worth noting that as more and more learners prefer this video-based learning experience, video sharing platforms have also become an online learning approach that cannot be ignored in addition to traditional MOOC platforms. Many high-quality open course videos have been uploaded to video sharing platforms such as YouTube so that more people can learn more easily (Burke & Snyder, 2008). Video sharing platforms with danmaku features are very popular with the younger generation. The term "danmaku" comes from Japanese and is essentially a system of video commentary that slides from one side of the screen to the other like a bullet (Howard, 2012). In China, most mainstream video platforms have danmaku functions, and the most popular platform is Bilibili, which has 54 million daily active users. In Europe, the TikTok platform which is popular among young people, some European streaming platforms and TV stations' web pages have also started to open the danmaku function in some shows.

However, with the popularity of online learning, some challenges have come to the fore, such as an increase in course dropout rates and a decrease in satisfaction with online courses, which may be related to learner interactions with others (Asoodar et al., 2016). The quality of interaction as a feature of online learning is also considered to be an important factor in improving the efficiency and effectiveness of online learning (Bernard et al., 2009). For learners, interaction through forums, discussion boards and danmaku in the online learning environment can promote cooperation and mutual support among different learners. Unlike traditional mainstream video platforms and MOOC platforms, video sharing platforms with Danmaku functionality offer a better user experience in terms of interactivity. In addition to interacting in the comments section, users can also comment instantly by sending a danmaku, which remains at the point in time of the current comment on the video being watched, and when other viewers play to that point in time, they will see other users' comments here. This sense of real-time commenting is difficult to experience on YouTube. As a result, many Chinese students choose to study online on video sharing platforms like Bilibili. Therefore, understanding learners' perceptions of danmaku in this particular context and exploring the application and value of danmaku can provide more options for MOOC improvement and in-class interaction for teachers, as well as provide some reference and help for the development of online learning platforms.

Based on this, the research questions for this study are as follows:

RQ1: How do users perceive the use of danmaku as a tool in open publicly shared MOOC videos?

RQ2: What are the specific roles of danmaku in open publicly shared MOOC videos?

On the basis of the Technology Acceptance Model (TAM), the authors combined the studies of other scholars to construct an analytical framework suitable for this study, which is to code and analyse the data with perceived usefulness, perceived ease of use and perceived enjoyment as the core structures (Davis 1989; Van der Heijden 2004; Rauniar et al. 2014; Zhang et al. 2014). Details will be presented in the next section.

In this study, content analysis was chosen as the research method. Content analysis is a systematic and objective quantitative research method to study information characteristics (Harwood & Garry, 2003). The content analysis method can perform quantitative coding analysis on objective barrage text data and perform in-depth qualitative analysis of the coding results. This study took the textual content of the danmaku as the research object, aiming to explore the acceptance and perception of danmaku by the users of video sharing platforms. These users mainly refer to students who study open publicly shared MOOC videos on video sharing platforms and learners who are interested in such videos. The data for this study were all collected from the official Bilibili platform using Python. The authors selected a total of 128 open publicly shared MOOC videos from 4 series of MOOC courses on the video sharing platform Bilibili, that is, Game Theory at Yale University, Positive Psychology at Harvard University, Macroeconomics at Peking University, and Mechanics of Materials at Tsinghua University. Two series of courses are taught in English with Chinese and English subtitles, and two are taught in Chinese with Chinese subtitles. Since the Bilibili platform assigns different weights to the danmaku content according to the user's level and behavior records, the higher the weight of the danmaku, the less likely it will be covered by the subsequent danmaku, which also means that the publisher of this part of the danmaku is more core user groups. Therefore, this study selected the first 100 danmaku of each MOOC video as the research sample, and used each individual danmaku text as an analysis unit, with 128 videos totaling 12174 analysis units. In this study, the qualitative analysis software MAXQDA 2022 was used as an analysis tool to analyze the content of danmaku. By combining the analysis framework mentioned in the previous section, the authors selected perceived usefulness, perceived ease of use and perceived enjoyment as the three primary indicators, and summarised the secondary indicators from the raw data through open coding. The coding process was carried out by two coders. To ensure objectivity, the author and another coder performed pre-coding after learning the key indicators and specific coding rules to test the consistency of their coding.

This study found that learners preferred using danmaku for instant messaging and interaction than traditional comment section discussions. Learners' perceptions of the use of danmaku as a learning tool in open publicly shared MOOC videos are mainly reflected in the areas of socialisation, information exchange, participation and interaction. The perceived usefulness of danmaku is mainly reflected in the fact that the use of danmaku is helpful for learners to obtain information and achieve effective communication. Perceived ease of use in this study refers to how easy and convenient it is for users to use the danmaku. The perceived enjoyment of using danmaku is reflected in the enjoyment and ease that users feel when using them, as the anonymous and open atmosphere created by danmaku video platform makes it easier for users to express their ideas. The main application scenarios of danmaku in open publicly shared MOOC videos are explanations, recommendations for videos and other related materials, explanations or corrections of subtitles, content discussions, communication conversations, simplified interaction, video interaction, comments, social interaction and emotional expression. These application scenarios also reflect the roles currently assumed by danmaku in the field of online learning. Danmaku adds a sense of ritual to MOOC learning, breaks the time and space limitations of teacher-student interaction and interaction between students, and transforms traditional MOOC learning into a decentralised online learning mode. Based on these unique and attractive features, more and more learners are choosing to use video sharing platforms with danmaku for their MOOC learning. The findings are useful for understanding the popularity of MOOC videos in video sharing platforms with danmaku features, the design of the danmaku function and the development of MOOC-based educational technology.

Asoodar M, Vaezi S, Izanloo B (2016) Framework to improve e-learner satisfaction and further strengthen e-learning implementation. Computers in Human Behavior 63:704–716. https://doi.org/10.1016/j.chb.2016.05.060 Bernard, R. M., Abrami, P. C., Borokhovski, E., Wade, C. A., Tamim, R. M., Surkes, M. A., & Bethel, E. C. (2009). A meta-analysis of three types of interaction treatments in distance education. Review of Educational Research, 79(3), 1243–1289. Burke, S. C., & Snyder, S. L. (2008). YouTube : An Innovative Learning Resource for College Health Education Courses An Overview of YouTube Using YouTube in Higher Education. International Electronic Journal of Health Education, 11, 39–46. Daradoumis, T., Bassi, R., Xhafa, F., & Caballé, S. (2013). A review on massive e-learning (MOOC) design, delivery and assessment. Proceedings - 2013 8th International Conference on P2P, Parallel, Grid, Cloud and Internet Computing, 3PGCIC 2013, 208–213. https://doi.org/10.1109/3PGCIC.2013.37 Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 319–340. Harwood, T. G., & Garry, T. (2003). An overview of content analysis. The Marketing Review, 3(4), 479–498. Howard, C. D. (2012). Higher order thinking in collaborative video annotations: Investigating discourse modeling and the staggering of learner participation. Indiana University. Rauniar, R., Rawski, G., Yang, J., & Johnson, B. (2014). Technology acceptance model (TAM) and social media usage: an empirical study on Facebook. Journal of Enterprise Information Management. Van der Heijden, H. (2004). User acceptance of hedonic information systems. MIS Quarterly, 695–704. Zhang, K. Z. K., Cheung, C. M. K., & Lee, M. K. O. (2014). Examining the moderating effect of inconsistent reviews and its gender differences on consumers’ online shopping decision. International Journal of Information Management, 34(2), 89–98.