Akman A. M., Hürses C., Yıldırım N., Gultekin-Karakas D.

30th International Conference of the International Association for Management of Technology: MOT for the World of the Future, IAMOT 2021, Cairo, Virtual, Egypt, 19 - 23 September 2021, pp.124-135 identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.52202/060557-0008
  • City: Cairo, Virtual
  • Country: Egypt
  • Page Numbers: pp.124-135
  • Keywords: Clustering, Digital transformation, Industry 4.0 policy, Policy analysis, Science and technology policy, Text mining
  • Middle East Technical University Affiliated: Yes


Copyright © 2021 by Naudé Scribante. Permission granted to IAMOT to publish and use.For policymakers, science and technology policy analysis is a critical competency that involves adopting various qualitative and quantitative techniques such as benchmarking, clustering, or content analysis. Traditional methods of manual systematic content analysis remain insufficient as they consume too much time-, bear the risk of using outdated data and embed the coding process's subjectivity. Results of qualitative content analyses are, therefore, often open to different interpretations. Text mining methods developed to test frequency, distribution, and co-occurrence of words could offer opportunities to analyse science and technology policies. However, in the context of digital transformation or Industry 4.0, text mining methods have been rarely utilized in the literature to provide insights into policymaking processes. This study primarily aims to present a text mining application to explore the country clusters by their Industry 4.0 and digital transformation policies and science and technology performances. The paper proposes a combined approach of science and technology performance evaluation and text mining clustering methods. It explores text mining methods' applicability as an alternative method for comprehensive analysis of national science and technology strategies. By integrating the performance indicators of science and technology to unsupervised learning techniques, we investigated whether the strategic country clusters match countries' science and technology performance-based clusters. We focused on locating Turkey within these clusters to present a concrete case application of the model and a case study for positioning the developing (follower) countries in the digital transformation strategic landscape.