Umeå University's logo

For automated documentation systems to be meaningful in pediatric rehabilitation, they must accurately capture and summarize information about a child's or youth's involvement in daily life activities. We present attention-ASP, a neuro-symbolic framework that combines Transformer attention with Answer Set Programming (ASP) to capture involvement-level information from interview-based text. By encoding domain-specific vocabularies as ASP programs, our model guides attention heads to focus on contextual cues such as place, time, activity, object, and people. Results show that symbolic reasoning over attention improves alignment with expert assessments, offering a promising direction for advancing clinical documentation tools in pediatric rehabilitation.

This paper addresses a conceptual and formal limitation in the International Classification of Functioning, Disability and Health (ICF), where activity and participation are treated as a unified construct, creating ambiguity in rehabilitation contexts. We propose an ontology keeping the ICF structure but formally separating these concepts, enabling more precise participation assessment. A case-based query analysis demonstrates how the unified ICF model may yield false positives in participation documentation, while the refined ontology supports nuanced evaluation. Implementation details and future evaluation are discussed.

The analysis of abrasive wear is central to sustainable material design and tool development, yet current practice relies on manual inspection of scanning electron microscopy (SEM) images, limiting scalability and reproducibility. We propose early work towards a neuro-symbolic approach that integrates convolutional neural networks for SEM image segmentation with an expert-elicited taxonomy of wear features encoded in Answer Set Programming. A curated dataset of 400 laboratory and field SEM images with expert-labeled annotations supports interpretable detection of wear mechanisms. This approach aims to reduce the dependency on large datasets, increase interpretability, in automated abrasive wear analysis. The contribution opens the way for scalable and transparent decision processes in tribology, with implications for efficient materials development and extended service life of industrial tools.

In this article, we introduce a general framework for structured argumentation providing consistent and well-defined justification for conclusions that can and cannot be inferred and there is certainty about them, which we call semantic and NAF-arguments, respectively. We propose the so-called semantic argumentation guaranteeing well-known principles for quality in structured argumentation, with the ability to generate semantic and NAF-arguments, those where the conclusion atoms are semantically interpreted as true, and those where the conclusion is assumed to be false. This framework is defined on the set of all logic programs in terms of rewriting systems based on a confluent set of transformation rules, the so-called Confluent Logic Programming Systems, making this approach a general framework. We implement our framework named semantic argumentation solver available open source.

Traditionally, in the argumentation theory literature structured arguments are constructed from rules interpretations aiming to build well-supported deductive evidence. Dierent from other approaches, we emphasize the role of investigating general frameworks that can also provide a consistent and well-defined justification for a conclusion that cannot be inferred and there is certainty about it, which we call here NAF-arguments, which have been less explored in the formal argumentation theory, despite its potential use in practical applications for building nuanced well-structured explanations and justifications.

This article introduces the so-called semantic argumentation guaranteeing well-known principles for quality in structured argumentation, and with the ability to generate two types of arguments, those where the conclusion atoms are semantically interpreted as true, and those where the conclusion is assumed to be false, we call them here semantic and NAF-arguments respectively. This framework is defined on the set of all logic programs in terms of rewriting systems based on a congruent set of transformation rules, the so-called Confluent Logic Programming Systems, making this approach a general framework. Additionally, we introduce a method for building such arguments using the program's strata through partial interpretations. We implement our framework named semantic argumentation solver available open source.

Building software agents that make decisions based on human values presents a complex challenge. This is a first attempt to systematically review trends in how researchers represent, select, evaluate, and implement values in software agents. The review highlights two key limitations in the field. First, most research adopts or acknowledges Schwartz’s theory of basic human values, but often simplifies its complexity. Second, many studies oversimplify the notion of values in general, failing to utilize established theories from psychology or philosophy. Additionally, the review finds that common software tools and existing frameworks, like the Belief-Desire-Intention model, are frequently used without fully considering their underlying principles. This research offers valuable contributions to the field of value-based decision-making in software agents, exposing research gaps and limitations in current research, and highlighting the need for a more nuanced approach. By calling for researchers to integrate insights from economics, social science, psychology, and neuroscience, the review paves the way for the development of more sophisticated agents that can navigate the complexities of real-world decision-making.

This paper presents a survey of software for computational argumentation, focusing on tools and libraries that can be re-used for building applications. The (so far preliminary) survey is intended as a continuously updated resource that researchers and potential practitioners can refer to when selecting tools for their argumentation use cases. The survey results indicate that while a plethora of argumentation tools exist, covering various formal argumentation approaches, the software engineering maturity of the tool ecosystem is low. For example, while many tools are openly available, few are properly documented and covered by automated tests. We argue that many of these shortcomings can be straightforwardly addressed by following simple engineering practices if community consensus regarding basic requirements for software artifacts can be established, as is the case in software engineering-oriented communities.

This study reviewed existing research on financial robo-advisors, focusing on the computational mechanisms from the artificial intelligence (AI) field, the methodological and ethical aspects of the proposed computational mechanisms, and the persuasive technology used in robo-advisors to convince or induce a specific financial activity in a person. We analyzed 48 financial robo-advisors, their algorithms (mainly machine learning) and their purposes (stock selection, user preference learning, etc.). We found a concerning lack of transparency regarding the data used to train these algorithms and a disregard for established trustworthiness criteria. Additionally, the persuasive design elements employed by robo-advisors were not well-explained in the reviewed literature. Notably, the study highlights the absence of social support principles (comparison, competition) commonly used in persuasive design. This work contributes to the financial AI community by suggesting better methodological approaches when AI-based persuasion is used.

Purpose: This study investigated the regulatory landscape surrounding artificial intelligence (AI) in the context of e-government development. The purpose of this paper is to identify record-keeping challenges, opportunities and weaknesses that emerge from AI loose regulation. The research focuses on Sweden, Finland and South Africa examining the interplay between existing guidelines, recommendations and legal structures at various levels.

Design/methodology/approach: The research adopted comprehensive systematic and scoping literature reviews, encompassing academic papers, reports and legal documents, along with an analysis of non-academic sources relevant to the study. This methodological approach helped to a get a deep understanding of the evolving AI regulatory frameworks.

Findings: There is currently limited research that focuses on the impact AI deployment has on the management of critical records in government administrations. Also, the findings reveal that AI regulatory environment varies from country to country. The European Union stands as a noteworthy example of a comprehensive framework for AI governance. In contrast, South Africa, while at its infancy stage, demonstrates potential initiatives and policies at different levels. There is emphasis on the need to focus on co-operation, skills development and uniform regulatory frameworks.

Practical implications: This research holds significant practical implications for policymakers, governmental bodies and stakeholders involved in AI governance. It emphasizes how crucial it is to incorporate AI alongside a solid records management system. The study advocates for strategic investments in education and skills development to enable individuals to navigate the complexities of AI governance.

Originality/value: This research adds to the existing body of knowledge by providing an examination of AI legislation in e-government in the context of public records management. The analysis helps to review literature and other research materials across different geographical areas. The study explores the distinctive strategies used by Sweden, Finland and South Africa. The recommendations offer policymakers and stakeholders suggestions on how to foster effective AI governance and innovation in the public sector but at the same time manage public records effectively.