MIPS - Mapping the Relationship between Research, Innovation, and Society through Topic Modeling

• Diogo Nolasco UFRJ
• Jonice Oliveira UFRJ

DOI: https://doi.org/10.5753/sbcas_estendido.2026.21030

Resumo

The proliferation of scientific data on the internet, including social networks, blogs, patents and articles, demands integrated analysis to guide strategic decisions. This data covers the life cycle of innovation, from scientific publications, through patents, to public perception of innovations. Understanding these dimensions allows you to better allocate resources and efforts in strategic areas that are well received by society. Although initiatives like Altimetry evaluate the specific impact of articles or authors, there is a gap in analyzing the interrelationships between ideas, discoveries, products and opinions over time. In this context, topic modeling emerges as a powerful tool for extracting and understanding themes in textual data. This work proposes an integrated technique to identify and analyze topics and their relationships in the scientific, technological and social spheres. The solution allows you to cross topics of different dimensions, revealing how an innovation is born, grows, turns into a commercial product and is perceived by society. Unlike current solutions that focus on a single dimension, this method offers a multidimensional view. The experiments carried out show the growth of research topics over time and how they were discussed by society, in addition to relating rumors and misinformation. The results indicate significant connections between the dimensions studied, highlighting the relationship of topics with milestones in the evolution of innovation. In the future, experiments will be conducted involving all three dimensions (science, technology and society), accompanied by the creation of visualizations to facilitate technological prospecting.

Referências

Blei, D. M., Ng, A. Y., and Jordan, M. I. (2003). Latent dirichlet allocation. Journal of Machine Learning Research, 3:993–1022.

Dresch, A., Lacerda, D. P., Jr, J. A. V. A., Dresch, A., Lacerda, D. P., and Antunes, J. A. V. (2015). Design Science Research, pages 67–102. Springer International Publishing.

Kullback, S. and Leibler, R. A. (1951). On information and sufficiency. The Annals of Mathematical Statistics, 22(1):79–86.

Nolasco, D. and Oliveira, J. (2020). Mining social influence in science and vice-versa: A topic correlation approach. International Journal of Information Management, 51.

Nolasco, D. and Oliveira, J. (2021). Topical rumor detection based on social network topic models relationship. sol.sbc.org.brD Nolasco, J OliveiraiSys-Brazilian Journal of Information Systems, 2021•sol.sbc.org.br, 14:5–27.

Nolasco, D. and Oliveira, J. (2025). Mirabr – a system for patent analysis. Simpósio Brasileiro de Sistemas de Informação (SBSI), pages 838–845.

World Health Organization (2020). Covid-19 advice - mythbusters — who western pacific. [link]. Accessed: 2026-02-23.