News big data analysis of international start-up innovation discourses through topic modelling and network analysis: comparing East Asia and North America

ABSTRACT

It is important to examine how start-up innovation is emerging in the information society. The media is a window through which innovation is highlighted in which fields. We conducted a comparative study with North America to comparatively analyze how startup innovation in East Asia appeared. This study used a set of computer analysis methods including Dirichlet-multinomial regression Topic Model and Topic network analysis. News articles from 2000 to 2019 were collected from East Asia and North America and were analysed on the topic of start-ups. The results indicated that the discourse of start-ups from East Asia and North America in the 2000s and 2010s showed distinctly different trends. The East Asia media changed its focus from an innovation economy to an emerging industry: mobility & energy, while the North American media showed a change from revenue sources to the benefit of innovation. In the discourse of start-ups in East Asia, the government-centred innovative economy was emphasised in the 2000s, and expectations for emerging industries such as mobility and energy increased after the 2010s. Meanwhile, in the discourse of start-ups in North America, revenue sources were emphasised in the 2000s, and the proportion of beneficiaries of innovation increased after the 2010s.

KEYWORDS:

• News discourse analysis
• start-up innovation
• text mining
• network analysis
• topic modelling

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Pointwise mutual information (PMI) is a measure of the interdependence between two variables in information theory. In this study, PMI was used to calculate the interdependence between two terms.

Additional information

Notes on contributors

Kyeo Re Lee

Kyeo Re Lee is a SW collaboration professor at Department of Software, Hanyang University ERICA, Ansan, South Korea. His research focuses on computational social sciences, human-AI interaction, metaverse and data science.

Jang Hyun Kim

Jang Hyun Kim is a professor at Department of Interaction Science, and a joint affiliate faculty at Department of Applied Artificial Intelligence, Sungkyunkwan University and his research focuses on social/semantic data analysis, social media, and future media.

Jaeyeon Jang

Jaeyeon Jang is a master at Department of Human-Artificial Intelligence Interaction, Sungkyunkwan University, Seoul, South Korea. Her reaserch intersts include data science, social network anlaysis and human-AI interaction.

Jeewoo Yoon

Jeewoo Yoon is a Ph.D. student at Department of Applied Artificial Intelligence in Sungkyunkwan University, Seoul, South Korea. His research interests include multimodal machine learning and affective computing.

Dongyan Nan

Dongyan Nan is a Ph.D. candidate at Department of Human-Artificial Intelligence Interaction/Department of Interaction Science, Sungkyunkwan University. His research focuses on Human-AI Communication, Consumer Experience, Business Informatics, Computational Social Science, Metaverse, and Network Analysis. He has authored 20 papers in major conferences and journals such as Technological Forecasting and Social Change, Information Processing and Managment, and Journal of Organizational and End User Computing.

Yonghwan Kim

Yonghwan Kim is working for Naver and PhD in Psychology. His research focuses on data science, metaverse, and social media.

Byungjun Kim

Byungjun Kim is a research associate in Center for Digital Humanities and Computational Social Sciences, Korea Advanced Institute of Science and Technology, Daejeon, South Korea. His research interests include digital humanities, computational social sciences, and natural language processing.