1. Mapping the evolution of social innovation in scientific publications : a topic modelling and text mining approachUroš Godnov, Jana Hojnik, Simona Kustec, 2025, izvirni znanstveni članek Opis: Objective: To trace how academic discourse on social innovation has evolved from 2000 – mid-2024 in numbers and leading topics by applying a special topic modelling and text mining methodology. Data & Sources: 4,703 full-text journal articles retrieved from Science Direct. Methods: Literature review and PDF text extracted with PyPDF2 and pdfplumber; cleaned and tokenised in R; topic modelling performed with Latent Dirichlet Allocation (ldatuning-optimised); temporal and correlation analyses visualised via tidyverse. Results: The number of publications increased significantly from 16 (in 2000) to 573 (in 2021), stabilizing thereafter. Seven dominant topics emerged: renewable energy, environmental/resource management, smart-city governance, sustainable food systems, corporate strategy, academic-method studies, and social-governance structures. “Social” and “innovation” became the top word pair after 2006; energy-related terms surged after 2016. Surprisingly, topics typically considered ‘social’ have not dominated the social innovation discourse in scientific communities compared to the aforementioned dominant topics. Discussion: Our results largely confirm existing findings from literature reviews and affirm the interdisciplinary, vague, contested, and still intensively evolving nature of social innovation. Dominant social innovation topics in scientific papers reference to social innovation topics in global political and policy documents, notably from the EU (from 2013 onwards) and the 2015 UN SDGs agenda, also emphasising collaboration between scientific, business, political and non-governmental stakeholders, and can thus serve as scientific, evidence-based advocacy for other stakeholders involved in social innovation processes. Conclusions: Social innovation research is now an established, systemic, and broadly interdisciplinary field of study, focusing on sustainability, emerging technologies, and governance topics. It is tightly connected with the political and policy agendas of leading international organisations, as well as business and non-governmental ones. Implications: Findings guide scholars to under-explored social-related content and niches (such as governance and, especially, equity topics) and help policymakers and other stakeholders involved in social innovation processes locate evidence-based approaches and clusters when designing their socially innovative responses, interventions, solutions, and measures.
Ključne besede: social innovation theories, global policy agenda, text mining, topic modelling, literature review
Objavljeno v DKUM: 05.09.2025; Ogledov: 0; Prenosov: 2
 Celotno besedilo (1,85 MB)
Gradivo ima več datotek! Več... |
2. Digital devices and interpersonal communication over timeTomaž Bratina, 2023, izvirni znanstveni članek Opis: Digital communication supported by mobile devices has an essential impact on interpersonal relations. Face-to-face communication is significantly affected because of the habits of simultaneous use of mobile devices and digital communication. The effect on interpersonal relations is negative since the behaviour affects personal closeness, empathy, and trust, including the feeling that a physically present person is unwanted or redundant, even replacing face-to-face communication with text messaging. The behavioural patterns significantly change if a loved one is involved. Results show that behaviour patterns did not change considerably over five years. Simultaneous digital communication, replacing face-to-face contact or voice call with text messaging is still present, with minor deviations in the post-pandemic period. Nevertheless, personal contact with beloved persons is still the primary preference over time compared to digital contact.
Ključne besede: mobile devices, interpersonal communication, simultaneous digital communication, phubbing, text messaging
Objavljeno v DKUM: 29.07.2025; Ogledov: 0; Prenosov: 3
Celotno besedilo (399,66 KB)
Gradivo ima več datotek! Več... |
3. |
4. AI model for industry classification based on website dataTimotej Jagrič, Aljaž Herman, 2024, izvirni znanstveni članek Opis: This paper presents a broad study on the application of the BERT (Bidirectional Encoder Representations from Transformers) model for multiclass text classification, specifically focusing on categorizing business descriptions into 1 of 13 distinct industry categories. The study involved a detailed fine-tuning phase resulting in a consistent decrease in training loss, indicative of the model’s learning efficacy. Subsequent validation on a separate dataset revealed the model’s robust performance, with classification accuracies ranging from 83.5% to 92.6% across different industry classes. Our model showed a high overall accuracy of 88.23%, coupled with a robust F1 score of 0.88. These results highlight the model’s ability to capture and utilize the nuanced features of text data pertinent to various industries. The model has the capability to harness real-time web data, thereby enabling the utilization of the latest and most up-to-date information affecting to the company’s product portfolio. Based on the model’s performance and its characteristics, we believe that the process of relative valuation can be drastically improved.
Ključne besede: industry classification, BERT transformer, business descriptions, multiclass text classification, AI
Objavljeno v DKUM: 01.07.2025; Ogledov: 0; Prenosov: 8
Celotno besedilo (1,01 MB)
Gradivo ima več datotek! Več... |
5. Thought experiments, fictions, and irrelevant detailsBojan Borstner, Tadej Todorović, 2025, izvirni znanstveni članek Opis: The article explores the problem of the cognitive value of thought experiments (TEs) and fictions. Specifically, it deals with the claim that fictions have cognitive value in virtue of being (elaborate) thought experiments. First, a short overview of the cognitive value of TEs is presented, followed by the recent findings from experimental philosophy, which cast doubt on the value of TEs. This is followed by an examination and rejection of the claim that fictions are TEs (as presented by Elgin) for two reasons. First, the analogy between scientific and thought experiments and fictions ultimately fails, as fictions contain the very variables that must be absent for performing successful scientific and thought experiments; second, because of this and based on the research in experimental philosophy, fictions should bias the reader to a greater degree than TE— this is shown to be collaborated by text comprehension research. This claim is further substantiated by analysing two examples of fictions, Le Guin’s The Matter of Seggri and her satirical piece A Modest Proposal: Vegempathy. Finally, a more modest claim is considered, namely that fictions contain TEs, which must be properly extrapolated and analysed, yet this leads to issues that are similar to the value of TEs debate. The article thus concludes that using TEs is not advisable for securing the cognitive value of fiction.
Ključne besede: thought experiments, fiction, experimental philosophy, cognitive value, text comprehension
Objavljeno v DKUM: 23.06.2025; Ogledov: 0; Prenosov: 3
Celotno besedilo (335,75 KB)
Gradivo ima več datotek! Več... |
6. Music with a Message : Words, Music and Propaganda 22025, zbornik Opis: Music with a Message: Words, Music and Propaganda 2 explores ways in which music in connection with various disciplines represents a voice of political critique, cultural expression and resistance. The chapters, written by established academics as well as graduate students, bring a plethora of fresh views from multiple fields like literature, language, English language teaching, as well as cultural, social and political studies. Musical genres addressed are as diverse as the approaches to propaganda; they expand from various ethnic folk and pop music, rap, alternative rock and metal to national anthems, cartoon soundtracks and musicals. This collection appeals to scholars, and students as well as enthusiasts interested in the profound cultural and ideological impact of music.
Ključne besede: literature, music, lyrics, text analyses, propaganda
Objavljeno v DKUM: 23.06.2025; Ogledov: 0; Prenosov: 2
Celotno besedilo (5,11 MB)
Gradivo ima več datotek! Več... |
7. The OpenScience Slovenia metadata datasetMladen Borovič, Marko Ferme, Janez Brezovnik, Sandi Majninger, Albin Bregant, Goran Hrovat, Milan Ojsteršek, 2020, drugi znanstveni članki Opis: The OpenScience Slovenia metadata dataset contains metadata entries for Slovenian public domain academic documents which include undergraduate and postgraduate theses, research and professional articles, along with other academic document types. The data within the dataset was collected as a part of the establishment of the Slovenian Open-Access Infrastructure which defined a unified document collection process and cataloguing for universities in Slovenia within the infrastructure repositories. The data was collected from several already established but separate library systems in Slovenia and merged into a single metadata scheme using metadata deduplication and merging techniques. It consists of text and numerical fields, representing attributes that describe documents. These attributes include document titles, keywords, abstracts, typologies, authors, issue years and other identifiers such as URL and UDC. The potential of this dataset lies especially in text mining and text classification tasks and can also be used in development or benchmarking of content-based recommender systems on real-world data.
Ključne besede: metadata, real world data, text data, text mining, text identification, natural language processing
Objavljeno v DKUM: 22.05.2025; Ogledov: 0; Prenosov: 8
Celotno besedilo (187,50 KB)
Gradivo ima več datotek! Več... |
8. Automatic classification of older electronic texts into the Universal Decimal Classification-UDCMatjaž Kragelj, Mirjana Kljajić Borštnar, 2021, izvirni znanstveni članek Opis: Purpose:The purpose of this study is to develop a model for automated classification of old digitised texts to the Universal Decimal Classification (UDC), using machine-learning methods.
Design/methodology/approach: The general research approach is inherent to design science research, in which the problem of UDC assignment of the old, digitised texts is addressed by developing a machine-learning classification model. A corpus of 70,000 scholarly texts, fully bibliographically processed by librarians, was used to train and test the model, which was used for classification of old texts on a corpus of 200,000 items. Human experts evaluated the performance of the model.
Findings: Results suggest that machine-learning models can correctly assign the UDC at some level for almost any scholarly text. Furthermore, the model can be recommended for the UDC assignment of older texts. Ten librarians corroborated this on 150 randomly selected texts.
Research limitations/implications: The main limitations of this study were unavailability of labelled older texts and the limited availability of librarians.
Practical implications: The classification model can provide a recommendation to the librarians during their classification work; furthermore, it can be implemented as an add-on to full-text search in the library databases.
Social implications: The proposed methodology supports librarians by recommending UDC classifiers, thus saving time in their daily work. By automatically classifying older texts, digital libraries can provide a better user experience by enabling structured searches. These contribute to making knowledge more widely available and useable.
Originality/value: These findings contribute to the field of automated classification of bibliographical information with the usage of full texts, especially in cases in which the texts are old, unstructured and in which archaic language and vocabulary are used.
Ključne besede: digital library, artificial intelligence, machine learning, text classification, older texts, Universal Decimal Classification
Objavljeno v DKUM: 28.01.2025; Ogledov: 0; Prenosov: 9
Celotno besedilo (1,91 MB)
Gradivo ima več datotek! Več... |
9. Text mining tourism literatureAjda Pretnar Žagar, Tomaž Curk, 2021, objavljeni znanstveni prispevek na konferenci Opis: Literature reviews are essential for understanding a specific domain as they map the main topics of current re-search. Our aim was to provide a framework for retrieving articles from online databases and analyzing them in a single script. We provide the analytical pipeline as open-source (https://github.com/tourism4-0/BibMine). The main research focus was on analyzing 318 abstracts from scientific papers on tourism and innovation, which we report in Zach et al. (2019). We used LDA topic modeling to uncover ten main topics, which we analyzed using pyLDAvis visualization. We used saliency and relevance scores to determine the main words that de-scribe a topic. The uncovered topics range from climate change and land use to smart destinations, travel expe-riences, and ICT. We performed similar analyses for the term "stakeholders," where we also observed the main verbs related to the query. Since verbs best define an activity, we used them to determine how stakeholders are involved in tourism development. Finally, we analyzed papers with the keyword "technology," where energy efficiency, VR, web technology, and augmented tourist experiences were the main topics.
Ključne besede: text mining, literature review, meta-analysis, topic modeling, tourism
Objavljeno v DKUM: 24.01.2024; Ogledov: 274; Prenosov: 7
Celotno besedilo (24,52 MB)
Gradivo ima več datotek! Več... |
10. Methodology of immersive video application : the case study of a virtual tourJure Jazbinšek, Gorazd Hren, 2021, izvirni znanstveni članek Opis: A Virtual Tour is an interactive presentation of real places accessible directly with an Internet browser with no additional installations of apps of plugins. Once, 360° photos are recorded and processed (stitched into spherical panoramas), editing of a Virtual Tour (walk) enables connection of spherical panoramic photos (or videos) into interactive presentations. For an enhanced experience and stand-alone presenting ability, features are added, like natural-sounding voice for text-to-speech descriptions and embedded videos. During multiple virtual tour presentations, users, viewers and presenters reported exceptional usability and an immersive experience. Virtual Tours have great potential to reshape the future education process and establish a new benchmark for presentation. The Virtual Tours application is expected to be used in education, tourism and future building sites or industry, as a key component for workforce briefings, and “as build” documenting of various stages of build, with the possibilities to integrate into Building Information Modelling (BIM) models.
Ključne besede: virtual tour, 360 camera, RICOH THETA Z1, 3dVista, Text to Speech
Objavljeno v DKUM: 13.11.2023; Ogledov: 443; Prenosov: 6
Celotno besedilo (2,21 MB)
Gradivo ima več datotek! Več... |
