© 2025 SURF
We present a novel architecture for an AI system that allows a priori knowledge to combine with deep learning. In traditional neural networks, all available data is pooled at the input layer. Our alternative neural network is constructed so that partial representations (invariants) are learned in the intermediate layers, which can then be combined with a priori knowledge or with other predictive analyses of the same data. This leads to smaller training datasets due to more efficient learning. In addition, because this architecture allows inclusion of a priori knowledge and interpretable predictive models, the interpretability of the entire system increases while the data can still be used in a black box neural network. Our system makes use of networks of neurons rather than single neurons to enable the representation of approximations (invariants) of the output.
LINK
Analyzing historical decision-related data can help support actual operational decision-making processes. Decision mining can be employed for such analysis. This paper proposes the Decision Discovery Framework (DDF) designed to develop, adapt, or select a decision discovery algorithm by outlining specific guidelines for input data usage, classifier handling, and decision model representation. This framework incorporates the use of Decision Model and Notation (DMN) for enhanced comprehensibility and normalization to simplify decision tables. The framework’s efficacy was tested by adapting the C4.5 algorithm to the DM45 algorithm. The proposed adaptations include (1) the utilization of a decision log, (2) ensure an unpruned decision tree, (3) the generation DMN, and (4) normalize decision table. Future research can focus on supporting on practitioners in modeling decisions, ensuring their decision-making is compliant, and suggesting improvements to the modeled decisions. Another future research direction is to explore the ability to process unstructured data as input for the discovery of decisions.
MULTIFILE
In this article, we calculate the economic impact of pilgrimage to Santiago de Compostela in the NUTS 2 region Galicia (Spain) in 2010. This economic impact is relevant to policymakers and other stakeholders dealing with religious tourism in Galicia. The analysis is based on the Input-Output model. Location Quotient formulas are used to derive the regional Input-Output table from the national Input-Output table of Spain. Both the Simple Location Quotient formula and Flegg's Location Quotient formula are applied. Furthermore, a sensitivity analysis is carried out. We found that pilgrimage expenditures in 2010 created between 59.750 million and 99.575 million in Gross Value Added and between 1, 362 and 2, 162 jobs. Most of the impact is generated within the 'Retail and Travel Services' industry, but also the 'Industry and Manufacturing', 'Services' and 'Financial and Real Estate Services' industries benefit from pilgrimage expenditures. This research indicates that in even in the most conservative scenario, the impact of pilgrimage is significant on the local economy of Galicia.
MULTIFILE
The circular economy (CE) is heralded as reducing material use and emissions while providing more jobs and growth. We explored this narrative in a series of expert workshops, basing ourselves on theories, methods and findings from science fields such as global environmental input-output analysis, business modelling, industrial organisation, innovation sciences and transition studies. Our findings indicate that this dominant narrative suffers from at least three inconvenient truths. First, CE can lead to loss of GDP. Each doubling of product lifetimes will halve the related industrial production, while the required design changes may cost little. Second, the same mechanism can create losses of production jobs. This may not be compensated by extra maintenance, repair or refurbishing activities. Finally, ‘Product-as-a-Service’ business models supported by platform technologies are crucial for a CE transition. But by transforming consumers from owners to users, they lose independence and do not share in any value enhancement of assets (e.g., houses). As shown by Uber and AirBNB, platforms tend to concentrate power and value with providers, dramatically affecting the distribution of wealth. The real win-win potential of circularity is that the same societal welfare may be achieved with less production and fewer working hours, resulting in more leisure time. But it is perfectly possible that powerful platform providers capture most added value and channel that to their elite owners, at the expense of the purchasing power of ordinary people working fewer hours. Similar undesirable distributional effects may occur at the global scale: the service economies in the Global North may benefit from the additional repair and refurbishment activities, while economies in the Global South that are more oriented towards primary production will see these activities shrink. It is essential that CE research comes to grips with such effects. Furthermore, governance approaches mitigating unfair distribution of power and value are hence essential for a successful circularity transition.
LINK
A B S T R A C T Background: Approximately 4 years ago a new concept of learning in practice called the ‘Learning and Innovation Network (LIN)’ was introduced in The Netherlands. To develop a definition of the LIN, to identify working elements of the LIN in order to provide a preliminary framework for evaluation, a concept analysis was conducted. Method: For the concept analysis, we adopted the method of Walker and Avant. We searched for relevant publications in the EBSCO host portal, grey literature and snowball searches, as well as Google internet searches and dictionary consults. Results: Compared to other forms of workplace learning, the LIN is in the centre of the research, education and practice triangle. The most important attributes of the LIN are social learning, innovation, daily practice, reflection and co-production. Often described antecedents are societal developments, such as increasing complexity of work, and time and space to learn. Frequently identified consequences are an attractive workplace, advancements of expertise of care professionals, innovations that endorse daily practice, improvement of quality of care and the integration of education and practice. Conclusions: Based on the results of the concept analysis, we describe the LIN as ‘a group of care professionals, students and an education representatives who come together in clinical practice and are all part of a learning and innovation community in nursing. They work together on practice-based projects in which they combine best practices, research evidence and client perspectives in order to innovate and improve quality of care and in which an integration of education, research and practice takes place’. We transferred the outcomes of the concept analysis to an input-throughput-output model that can be used as a preliminary framework for future research.
DOCUMENT
More than 25!years after Moore’s first introduction of the public value concept in 995, the concept is now widely used, but its operationalization is still considered difficult. This paper presents the empirical results of a study analyzing the application of the public value concept in Higher Education Institutions, thereby focusing on how to account for public value. The paper shows how Dutch universities of applied sciences operationalize the concept ‘public value’, and how they report on the outcome achievements. The official strategy plans and annual reports for FY2016 through FY2018 of the ten largest institutions were used. While we find that all the institutions selected aim to deliver public value, they still use performance indicators that have a more narrow orientation, and are primarily focused on processes, outputs, and service delivery quality. However, we also observe that they use narratives to show the public value they created. In this way this paper contributes to the literature on public value accounting.
DOCUMENT
