Érudit's RSS Feed

A Quantum Leap for Classification: Toward a New Framework for Library Science Beyond Dewey

Abderrahmane Kalem
The classical paradigm of knowledge organization, epitomized by hierarchical systems like Dewey Decimal Classification (DDC), is predicated on a nineteenth-century model of certainty and order. Contemporary scholarship has revealed its inherent limitations, including structural biases, epistemic injustice, and an inability to accommodate the fluid, interdisciplinary nature of modern knowledge. While computational models have increased efficiency, they largely inherit the classical goal of reducing ambiguity to a single, authoritative outcome. This paper argues for a paradigm shift, proposing a novel theoretical framework for classification inspired by the field of quantum cognition. The framework reconceptualizes the information object not as a static entity with intrinsic properties, but as a dynamic system of potential meanings. We adapt three core concepts from quantum theory for library science: Knowledge Superposition to model interdisciplinarity, Semantic Entanglement to provide a formal basis for serendipity, and Observer-Dependent Measurement to address issues of subjectivity and classificatory authority. The viability of this framework is demonstrated through a computational proof-of-concept that simulates how a single, multifaceted document can yield different, probabilistic classifications based on the user’s contextual perspective. This approach offers a robust solution to long-standing challenges, shifting classificatory authority from the institution to the user and providing a path toward transforming the library from a static repository into a dynamic, participatory knowledge system.

Generative AI for automatic topic labelling

Diego Kozlowski, Carolina Pradier and Pierre Benz
Topic modelling has become a prominent tool for the study of scientific fields, as they allow for a large-scale interpretation of research trends. Nevertheless, the output of these models is structured as a list of keywords, which requires a manual interpretation for the labelling. This paper proposes to assess the reliability of three LLMs, namely flan, GPT-4o, and GPT-4 mini for topic labelling. Drawing on previous research leveraging BERTopic, we generate topics from a dataset of all the scientific articles (n=34,797) authored by all biology professors in Switzerland between 2008 and 2020, as recorded in the Web of Science database. We assess the output of the three models both quantitatively and qualitatively and measure the effect of the temperature parameter in GPT models and find that, first, both GPT models are capable of correctly and precisely labelling topics from the models' output keywords at the default temperature. Second, 3-word labels are preferable to grasp the complexity of research topics.

Intergenerational Digital Storytelling for Inuvialuit Cultural Heritage Preservation

Rebecca Sasso Annunziata, Sharon Farnel and Ali Shiri
This paper provides an overview of mobile storytelling applications developed for and by Indigenous communities. It identifies and examines a variety of mobile applications to support Indigenous cultural and linguistic heritage. These applications include those reported in the literature as well as those available in application stores. The review provides a detailed overview of the mobile application interface functions and features, including recording capabilities, language-related features, accessibility, content categorization, and geolocation. Based on the review, this paper provides a set of design recommendations for developing mobile applications for Indigenous communities, including for Indigenous protocols, voices, and community needs.