Our Work

This study investigates the ability of Large Language Models (LLMs) to automatically index educational resources. To this end, a case study was conducted involving ChatGPT Plus, DeepSeek, and Gemini, …

This study investigates the ability of Large Language Models (LLMs) to automatically index educational resources. To this end, a case study was conducted involving ChatGPT Plus, DeepSeek, and Gemini, comparing their performance to that of academic authors and librarians. The methodological approach includes the adoption of matching criteria that allowed for the establishment of different levels of coverage and progression. The data were analyzed using precision, coverage, and F-measure metrics. The models produced results similar to the keywords assigned by the authors. On the other hand, the correspondence with the descriptors included by librarians showed the lowest match. In other words, the LLMs were closer to derived indexing (direct extraction from the document) than to indexing by attribution (artificial language), as employed by librarians. It is concluded that, based on the results, these models are capable of performing semantic indexing. However, there is difficulty in reproducing the standardized, institutional, or technical logic of library indexing, making it necessary to expand the studies.Conference abstract published in the proceedings of TOI2026 — XII Congresso Internacional em Tecnologia e Organização da Informação. ISSN 2965-4130.Original publication in Doity

  Twelve reviews published between 2011 and 2025 were examined. The six with an enumerable primary-study corpus were analysed for evidence overlap using the corrected covered area, for the thresh…

  Twelve reviews published between 2011 and 2025 were examined. The six with an enumerable primary-study corpus were analysed for evidence overlap using the corrected covered area, for the thresholds they report and the basis on which those were derived, and for the strength of their practice recommendations relative to their appraised methodological confidence. This repository contains the study-by-review incidence matrix, the index of unique primary studies, the analysis plan, a dated log of amendments to it, the claim-strength coding files of both raters, and a single script that recomputes every figure reported in the manuscript. The analysis plan was written and dated before extraction but was NOT prospectively registered; it is deposited here unaltered together with the amendment log. All source material analysed is published literature and is cited in full in the manuscript. No individual-level data are included.

This article analyzes the ethical challenges arising from the use and manipulation of images by digital and print media, considering their role as historical records and documentary evidence. It start…

This article analyzes the ethical challenges arising from the use and manipulation of images by digital and print media, considering their role as historical records and documentary evidence. It starts from the premise that modern editing technologies have eroded the concept of "photographic truth." The focus is therefore to investigate how ethics can respond to this technological escalation to ensure the validation of images, whose intrinsic vulnerability calls into question manipulation practices that are dangerously normalized and socially accepted. The methodology combines bibliographic and documentary research with the analysis of emblematic case studies in which image manipulation was central to constructing false narratives. The main conclusion highlights that the popularization of smartphones and their editing features has amplified the use of images, establishing them as one of the most universal forms of communication in contemporary society. In this context, the figure of an "ethical mediator" is proposed as an alternative for restoring the credibility of images, given the risk that this resource may become a source of suspicion rather than certainty. Finally, the study reinforces the importance of Information Ethics and the need to develop mechanisms that ensure a safer informational environment in the infosphere.Conference abstract published in the proceedings of TOI2026 — XII Congresso Internacional em Tecnologia e Organização da Informação. ISSN 2965-4130.Original publication in Doity

This study presents Mapping and Planning for Developmental Inclusion (MAP-DI) as a longitudinal informational infrastructure for Inclusive Special Education. Grounded in Information Science, inclusive…

This study presents Mapping and Planning for Developmental Inclusion (MAP-DI) as a longitudinal informational infrastructure for Inclusive Special Education. Grounded in Information Science, inclusive education, neuroethics, educommunication, user experience, and technology adoption, the proposal addresses the fragmentation of pedagogical, family, curricular, functional, and orientative tele-screening records. MAP-DI is non-diagnostic: it does not automate eligibility, replace professional judgment, or classify students through clinical inference. Instead, it organizes, retrieves, contextualizes, and communicates information to support human, ethical, documented, and reviewable pedagogical decisions, strengthening case studies, Specialized Educational Assistance Plans, Individualized Educational Plans, and inclusive school governance.Conference abstract published in the proceedings of TOI2026 — XII Congresso Internacional em Tecnologia e Organização da Informação. ISSN 2965-4130.Original publication in Doity

This extended abstract presents the Theory of Information Framework connected to the PDSGOEDAL+ vocabulary as a basis for semantic tagging, knowledge organization and information retrieval. The propos…

This extended abstract presents the Theory of Information Framework connected to the PDSGOEDAL+ vocabulary as a basis for semantic tagging, knowledge organization and information retrieval. The proposal assumes that people, documents, systems, groups, organizations, events, devices, living beings, places, processes, subjects, objects, tools, ideas, motivations and questions can be treated as interacting informational entity classes. Each entity may receive a unique identifier, minimum metadata, contextual relations and semantic layers. The aim is to discuss how this architecture may support libraries, archives, repositories, personal knowledge systems and artificial intelligence applications. The main contribution is an initial language of entities and relations that can improve interoperability between human and computational systems while preserving context, authorship, provenance and traceability.Conference abstract published in the proceedings of TOI2026 — XII Congresso Internacional em Tecnologia e Organização da Informação. ISSN 2965-4130.Original publication in Doity

This extended abstract discusses hybrid societies composed of human beings and artificial intelligence agents. Generative systems, software agents, robots, drones and humanoid devices are no longer me…

This extended abstract discusses hybrid societies composed of human beings and artificial intelligence agents. Generative systems, software agents, robots, drones and humanoid devices are no longer merely external tools: they participate in information networks, institutional workflows, scientific communication, legal drafting, education, decision support and actions in the physical world. The classical Turing Test is therefore insufficient. The central question is not only whether a machine can imitate human conversation, but how human and non-human agents jointly produce texts, records, decisions, interactions and consequences. The research problem is what academic and research communities should do when AI systems become active participants in knowledge production and circulation. The proposal connects this debate to authorship, provenance, prompt transparency, levels of AI participation, human accountability and public auditability.Conference abstract published in the proceedings of TOI2026 — XII Congresso Internacional em Tecnologia e Organização da Informação. ISSN 2965-4130.Original publication in Doity

This extended abstract discusses the authorship limbo created by the use of artificial intelligence in academic, scientific, legal and technological writing. The central thesis is that current initiat…

This extended abstract discusses the authorship limbo created by the use of artificial intelligence in academic, scientific, legal and technological writing. The central thesis is that current initiatives on AI disclosure are important but still fragmented, not fully integrated into publication systems and not yet present everywhere. A new gray literature may be emerging inside formal literature: texts produced with relevant AI participation but without clear declaration of use, prompts, levels of contribution or human accountability. The research problem is what should be done by journals, conferences, universities, courts, public agencies and organizations. The proposal argues for explicit AI-use declarations in all publications, levels or seals of AI participation, prompt and provenance transparency, public policies and responsible human authorship. The contribution is to separate technical risks, such as hallucinations and quality, from the deeper problem of authorship, responsibility and informational traceability.Conference abstract published in the proceedings of TOI2026 — XII Congresso Internacional em Tecnologia e Organização da Informação. ISSN 2965-4130.Original publication in Doity

The dataset consist details of respondents measuring their perceptions on constructs from the Diffusion of Innovation theory and metacognitive regulation.

How are agrifood systems changing in response to new nutritional, environmental, and market needs? And what role can new protein sources play in this transformation? The event is promoted by the CIRI …

How are agrifood systems changing in response to new nutritional, environmental, and market needs? And what role can new protein sources play in this transformation? The event is promoted by the CIRI Agri-Food of the University of Bologna and will take place on October, 23rd 2026. The event brings together demographic, economic, technological, nutritional, and consumer behaviour perspectives to offer an integrated view of the transformations underway. At the heart of the discussion: population and consumption: demographic scenarios, new dietary models, and growing pressure on resources; markets and supply chains: evolving demand, new value chains, and emerging business models; technological innovation: plant-based, fermentation, cell cultures, new ingredients, and hybrid products; nutrition and sustainability: nutritional quality, environmental impacts, and new challenges for the food system; consumers and the market: acceptability, perceptions, and the factors that guide food choices; future opportunities: prospects for businesses, researchers, policy makers, and institutions. Through scientific contributions and a roundtable with companies, the workshop brings research and the market together to understand which proteins we will eat, how they will be produced, and what opportunities they may generate for the agri-food system.

As the semiconductor industry rapidly expands, there is a need for new development methods, especially in the domain of safety-critical designs. The ISO 26262 standard plays a crucial role in the auto…

As the semiconductor industry rapidly expands, there is a need for new development methods, especially in the domain of safety-critical designs. The ISO 26262 standard plays a crucial role in the automotive industry by requiring systems to operate safely and reduce the chance of critical failures. Advanced, automated approaches are essential to meet these challenges effectively. This paper presents an automated framework for safety verification using the principles of model-driven architecture, which aims to increase the efficiency, quality, and trustworthiness of these systems. It introduces the concept of creating models with different levels of granularity for various components of a design such as gate-level models for the safety-critical parts and Register-Transfer Level (RTL) models for the rest of the designs. The innovation includes adding fault injectors to these models to inject faults directly in the design. Through a holistic generation flow, various safety verification techniques have been explored, including methods based on simulation, emulation, and formal verification. The effectiveness of these methods has been demonstrated in many different industrial desion applications.