In recent years, the question of the reliability of Machine Learning (ML) methods has acquired significant importance, and the analysis of the associated uncertainties has motivated a growing amount of research. However, most of these studies have applied standard error analysis to ML models—and in particular Deep Neural Network (DNN) models—which represent a rather significant departure from standard scientific modelling. It is therefore necessary to integrate the standard error analysis with a deeper epistemological analysis of the possible differences between DNN (...) models and standard scientific modelling and the possible implications of these differences in the assessment of reliability. This article offers several contributions. First, it emphasises the ubiquitous role of model assumptions (both in ML and traditional science) against the illusion of theory-free science. Secondly, model assumptions are analysed from the point of view of their (epistemic) complexity, which is shown to be language-independent. It is argued that the high epistemic complexity of DNN models hinders the estimate of their reliability and also their prospect of long term progress. Some potential ways forward are suggested. Thirdly, this article identifies the close relation between a model’s epistemic complexity and its interpretability, as introduced in the context of responsible AI. This clarifies in which sense—and to what extent—the lack of understanding of a model (black-box problem) impacts its interpretability in a way that is independent of individual skills. It also clarifies how interpretability is a precondition for a plausible assessment of the reliability of any model, which cannot be based on statistical analysis alone. This article focuses on the comparison between traditional scientific models and DNN models. However, Random Forest (RF) and Logistic Regression (LR) models are also briefly considered. (shrink)

This study maps the evolution of research themes on datafication, analyzing trends, key authors, interdisciplinary collaborations, and emerging topics from 1994 to 2023. The analysis reveals a notable increase in publication volume, particularly from 2014 onwards, reflecting advancements in digital technologies and heightened interest in data-driven research. A significant surge occurred during the COVID-19 pandemic, with 26.10% of total publications in 2022 and 30.52% in 2023 alone. Thematic clusters identified through keyword mapping include Social Media and Privacy, Artificial Intelligence and (...) Machine Learning, Human Dimensions, and Infrastructure and Trust, highlighting diverse research foci. Emerging discussions on data justice and inequality reflect growing attention to the ethical and socio-political implications of datafication. The study also examines the types of documents and subject areas, revealing the dominance of peer-reviewed journal articles (71.41%) and a strong representation of social sciences (46.93%), computer science (14.75%), and arts and humanities (11.57%). Interdisciplinary connections underscore the broad impact of datafication across technology, healthcare, education, and media studies. This research offers insights into the dynamic nature of datafication, pointing to the need for further interdisciplinary collaboration, especially in addressing societal and ethical concerns such as data governance and digital inequality. Future research directions should focus on the human dimensions of datafication, data literacy, and the development of robust data governance frameworks to mitigate potential inequalities and power imbalances in a rapidly data-driven world. (shrink)