Characteristics of Chinese Online Movie Reviews and Opinion Leadership Identification

ABSTRACT

This article uses a case study of the reception of popular South Korean films in China to investigate aspects of proactive participation in the Chinese social mediasphere. Based on the big data collected from the popular Douban online review platform, the authors provide an in-depth investigation of the characteristics of online opinion leaders and leadership. First, we statistically analyze user-generated content-related features for both leader and non-leader groups. In addition, we employ a Bayesian optimization-based imbalanced learning algorithm to identify characteristics of the minority opinion leaders. The efficiency of the proposed algorithm is evaluated based on 17,786 reviews collected from 35 popular films covering a variety of genres, and the proposed algorithm is shown to perform accurately by identifying opinion leaders, compared to other state-of-the-art approaches. Our study also reveals some major features associated with opinion leaders, such as vocabulary, viewing habits, and friendship networks. Finally, we show how these elements are contributing to new understandings of the rapidly transforming digital domain of Chinese social media and the roles of such platforms.

Acknowledgments

This article draws on research conducted in association with an Australian Research Council Discovery Project, Digital China: From Cultural Presence to Innovative Nation (DP-170102176).

We also thank the anonymous reviewers for their valuable comments on a previous draft of this article, in particular with regard to the statistical test.

Notes

1. In comparison, registered users that are active in Douban’s short reply section can post messages that are a maximum of 140 Chinese characters, which by our count is equivalent to around 500 English characters.

2. www.siteworthtraffic.com/report/douban.com. Accessed 19 April 2017.

3. Conventional abbreviations include: SBV = subject of verb; VOB = object of verb; IOB = indirect object; FOB = fronting object; DBL = double roles: subject & object; ATT = attribute; ADV = adverbial; CMP = complement; COO = coordinate; POB = preposition-object; LAD = left adjunct; RAD = right adjunct. More details about the dependency types can be found in http://ltp.readthedocs.io/zh_CN/latest/appendix.html#id5.

4. We employ an open-source library for sentence segmentation, and its detail can be found in https://github.com/fxsjy/jieba.

5. Accessible from https://movie.douban.com/review/5150867.

Additional information

Funding

This work was supported by the Australian Research Council Discovery Project, Digital China: From Cultural Presence to Innovative Nation [ARC DP-170102176].

Notes on contributors

Jie Yang

Jie Yang is Lecturer in Big Data and Database from School of Computing and Information Technology at the University of Wollongong. His research involves Machine Learning and (Big) Data Mining. Dr Yang has contributed significantly to numerous cross-discipline projects, on the topic of social media and health care.

Brian Yecies

Brian Yecies is Associate Professor in Communication and Media at the University of Wollongong, where he researches on film, digital media, creative industries, innovation ecosystems, and cultural policy. He is a chief investigator on two Australian Research Council Discovery Projects.

Peter Yong Zhong

Peter Yong Zhong is Research Assistant in Communication and Media, School of the Arts, English and Media, Faculty of Law, Humanities and the Arts at the University of Wollongong. He is working with Jie Yang and Brian Yecies on two prestigious Australian Research Council Discovery Projects.