Exploring Technoscience in the Public Sphere: Opportunities through Open Web Search

EXPLORING TECHNOSCIENCE IN THE PUBLIC SPHERE: OPPORTUNITIES THROUGH OPEN WEB SEARCH E. Di Buccio∗, University of Padova, Padova, Italy Abstract Interdisciplinary research fields, such as Science and Technology Studies (STS), investigate the relationship between science, technology, and society, and how they influence each other. These investigations can benefit from the extraction and the analysis of different representations of science and technology issues — for example, comparing the perspectives on Nuclear Power of social media users or newspaper readers with those of scientists and other experts. The vast amount of digitized content available online enables the use of diverse sources for these representations, thus fostering the opportunity to consider a plurality of perspectives. Examples of digitized informative resources include various media streams, such as social media posts and online newspaper articles. These types of sources are currently utilized in the TIPS project [6], an interdisciplinary initiative specifically aimed at studying science and technology discourse within the public sphere. To support researchers in fields such as the social sciences, a dedicated web platform has been designed and developed [2]. The platform is built on open-source libraries to promote reproducibility and is structured using a service-oriented architecture. In addition to media streams, other web-based informational resources may serve to complement existing representations. As noted by Lewandowski [4], an Open Web Index can be a valuable asset for researchers across various disciplines, including Computational Social Science. In projects such as TIPS, the primary users are experts from research fields such as the Social Sciences, Humanities, and Communication Studies. These users often require access to and analysis of specific segments of the Web to support their investigations. This need aligns with the objectives outlined in [3] and the Open Web Search project, which aims to make the index openly accessible as data. Moreover, selected portions of this index may serve as the foundation for developing specialized search engines—such as the one designed and implemented in the TIPS project to assist researchers in these domains. Those portions of the index might need to be enriched with specific metadata necessary for the experts analysis. Examples of such metadata include the “actors” mentioned in the informative resources, e.g., named entities; these are examples mentioned in [3] in the “semantic enrichment” step. However, other indicators are useful to investigate specific research questions in the considered application scenario. Examples of those indicators might be those devised to measure the degree to which a semantic dimension is present in a document, e.g., the presence of“risk”; those indicators might rely on controlled vocabularies [1] or more ∗[email protected] complex techniques relying on embedding-based representations. Other metadata might include linguistic properties or related measures, e.g., readability. Because of the efficiency constraints on processing huge amount of data, those metadata might be available only for specific portions and, for instance, accessible through specific verticals. Furthermore, within the scope of a single research project, multiple verticals may be employed to address the diversity of public discourse sources. For example, in the case of the TIPS project, dedicated verticals can be developed to target specific types of content, such as online news, forums or scientific blogs. Drawing on the extensive body of literature in resource selection, particularly from the domains of Distributed Information Retrieval (IR) and Federated Web Search [5], future versions of the platform can not only leverage portions of the open web index, but also rank these portions according to their relevance to particular technoscientific issues. Resource selection algorithms might be also be the basis for novel indicators to support expert analysis, e.g., for measuring the prominence of technoscientific issue in the public discourse. REFERENCES [1] Di Buccio, E., Lorenzet, A., Melucci, M., Neresini, F. (2016). Unveiling latent states behind social indicators. CEUR Workshop Proceedings, 1831. [2] Di Buccio, E., Cammozzo, A., Neresini, F., Zanatta, A. (2022). TIPS: Search and Analytics for Social Science Research. CEUR Workshop Proceedings, 3178. [3] Granitzer, M., Voigt, S., Fathima, N. A., Golasowski, M., Guetl, C., Hecking, T., Hendriksen, G., Hiemstra, D., Martinovič, J., Mitrović, J., Mlakar, I., Moiras, S., Nussbaumer, A., Öster, P., Potthast, M., Srdič, M. S., Megi, S., Slaninová, K., Stein, B., Zerhoudi, S. (2024). Impact and development of an Open Web Index for open web search. Journal of the Association for Information Science and Technology, 75(5), 512–520. https://doi.org/10.1002/asi.24818 [4] Lewandowski, D. (2019). The web is missing an essential part of infrastructure. Communications of the ACM, 62(4), 24–24. https://doi.org/10.1145/3312479 [5] Shokouhi, M., Si, L. (2011). Federated Search. Foundations and Trends in Information Retrieval, Volume 5, Issue 1, 1–102. https://doi.org/10.1561/1500000010 [6] https://www.tipsproject.eu/tips [Last visited: April 30, 2025] https://doi.org/10.5281/zenodo.17238455