Intelligence and Equity: Shaping the Future of Knowledge

Automatic Subject Indexing (ASI) plays a critical role in knowledge organization and information retrieval. However, existing studies remain fragmented, often emphasizing algorithmic details while overlooking the broader intellectual structure and thematic evolution of the field. To address this gap, this study constructs a comprehensive knowledge map of ASI research spanning 2000–2024 by integrating scientometric assessment with content analysis. Based on 1,161 publications retrieved from Web of Science and Scopus, we applied co-authorship, co-citation, and thematic evolution analyses to reveal influential contributors, collaboration patterns, and topic dynamics. To complement the scientometric insights, 26 representative studies on technical innovations and domain applications were examined in depth. Results reveal three major developmental phases: an initial reliance on rule-based approaches, a subsequent shift toward statistical and machine learning techniques, and the current dominance of deep learning architectures. Within this latest phase, large language models (LLMs) have emerged as a transformative development, while recent research also underscores multilingual indexing, ontology alignment, and linked data integration as critical directions for improving semantic interoperability, knowledge organization, and retrieval effectiveness. Despite progress, persistent challenges include data scarcity for low-resource languages, limited alignment between algorithmic outputs and professional cataloging standards, and the opacity of deep learning models. Key opportunities include leveraging LLMs for context-aware indexing, advancing human-in-the-loop workflows, and promoting open datasets and benchmarks to foster transparency and interoperability. This study offers an integrated analysis of ASI research, providing structural insights, critical interpretation of methodological trends, and forward-looking perspectives for scholars and practitioners.

This study aims to construct a multi-stage rumor detection framework based on Retrieval-Augmented Generation (MS-RAG-RD) that enhances the accuracy and credibility of rumor detection. The MS-RAG-RD model achieved an F1 score of 91.3% and a robustness score of 94.97 for rumor detection performance. The model exhibited strong resistance to interference and low vulnerability to structured semantic perturbations in adversarial attacks. Ablation experiments validated the individual module contributions, emphasizing the critical role of the hybrid retrieval module in achieving performance gains. The study results indicated that the MS-RAG-RD model could enhance the performance of the rumor detection task, providing a technical framework for real-time monitoring and accurate tracing of misinformation in cyberspace governance.

Large Language Models (LLMs) are increasingly utilized in digital libraries and knowledge systems to facilitate access to cultural and historical information. However, their outputs can reproduce subtle biases, particularly when addressing minority and low-resource communities. This study evaluates seven state-of-the-art LLMs on ten English prompts that embed culturally sensitive and potentially biased assumptions related to Vietnam, Myanmar, and Nepal. We systematically analyze these prompts’ responses for the presence of subtle bias, including gender stereotyping, linguistic ethnocentrism, epistemic bias, victim-blaming, and cultural essentialism. Our findings reveal significant variation in bias prevalence and type across models, with some exhibiting pervasive stereotyping and cultural marginalization, while others demonstrate more balanced and nuanced responses. These results emphasize the necessity for robust bias mitigation, culturally diverse training data, and human-in-the-loop oversight when deploying LLMs in digital heritage contexts. We discuss implications for ethical AI development in knowledge access and outline directions for future research to ensure fairness, transparency, and inclusivity in culturally sensitive AI applications.

Due to growing concerns about the risks associated with artificial intelligence (AI), interest in trustworthy AI (TAI) has amplified. Currently, few empirical studies investigate (a) whether TAI factors indeed encourage individuals to increase their usage of Generative AI (GenAI), compared to motivators such as those identified in the Unified Theory of Acceptance and Use of Technology (UTAUT) and (b) whether users’ levels of motivation from specific TAI and UTAUT factors vary with their GenAI usage types (e.g., obtaining answers, assisting with writing). Such research focusing specifically on Asia-Pacific nations, such as Singapore, is even rarer. This study thus conducted an online survey of 300 adults in Singapore to explore the two research gaps above. Data were analyzed using descriptive and inferential statistics (multiple regressions). The study found “Effort Expectancy” and “Performance Expectancy” (both from UTAUT) to be the top motivators for GenAI usage, followed by the “Technical Robustness and Safety” TAI. Significant associations were observed between types of GenAI usage and motivation levels from different TAI and UTAUT factors. For instance, using GenAI “to get answers” was correlated with being motivated to increase GenAI usage by “Technical Robustness and Safety”, and using GenAI “to solve problems” was associated with being encouraged by the “Human Agency and Oversight” TAI. Among the demographic variables, education yielded the most statistically significant associations (five out of 11). The implications for AI governance, system design, and stakeholder engagement and training are discussed.

Generative Artificial Intelligence (AI) models, especially large language models (LLMs), are increasingly used to retrieve and generate information in digital libraries. However, these models often reflect cultural biases and stereotypes that distort or marginalize knowledge representations. This paper tackles bias in LLM-generated English text on Asian history and culture. We formally define bias categories, including stereotyping, omission, ethnocentrism, and simplification, in the context of generative AI outputs. We propose a novel framework combining multi-perspective generation with bias detection to mitigate such biases. Supported by a theoretical analysis, we introduce formal bias measures and prove that under ideal conditions, our method can eliminate stereotypical content and perspective omissions. Furthermore, we present a bias annotation scheme and algorithm that generates answers incorporating diverse cultural viewpoints while filtering out identified stereotypes. Our approach provides formal guarantees for bias reduction, advancing the state-of-the-art by bridging bias mitigation, information retrieval, and digital library research to promote fairness and cultural inclusivity in AI-generated content.

Detecting sarcasm in Chinese texts presents unique challenges due to its indirect expression, which complicates accurate identification. Inaccurate assessments of sarcastic content on social media platforms often lead to negative user interactions, highlighting the importance of precise sarcasm detection. Emojis, which are widely used on Chinese social networking platforms such as Sina Weibo (hereafter Weibo), a service similar to X (formerly Twitter) and Facebook, add additional layers to textual communication, conveying emotions, intentions, and potentially sarcasm. However, the lack of well-annotated, high-quality Chinese datasets poses a significant obstacle to effective sarcasm detection, while the contextual complexity of Chinese sarcasm remains a major challenge for current language models. To address these issues, we propose a method that integrates four distinct modules to achieve comprehensive sarcasm detection in Chinese social media comments, particularly short user-generated texts with emoji interactions. Our model leverages emojis as a critical feature and capitalizes on the structural and lexical characteristics of Chinese sarcastic sentences. By incorporating features from both emoji-enhanced and plain text representations, the model demonstrates significantly improved accuracy in detecting sarcasm. Additionally, to support the training, testing, and validation of the system, we constructed a carefully designed dataset comprising multiple subsets. These subsets not only support the model’s training and evaluation but also serve as valuable resources for future research on Chinese sarcasm detection.

This paper proposes an interactive museum-viewing support system designed to foster diverse perspective-taking through role-play with fictional characters. While museums are widely regarded as valuable informal learning environments, passive viewing often results in low engagement and limited knowledge retention. Inspired by Visual Thinking Strategies (VTS), our tablet-based application leverages large language models (LLMs) to generate context-sensitive, character-driven questions posed by fantasy personas such as elves, dwarves, and werewolves, each with distinctive values and interpretive tendencies. Visitors engage with these questions and are occasionally prompted to assume a character’s role, generating their own questions in that persona’s voice. To explore the feasibility and user response, we conducted a small-scale case study at the National Museum of Ethnology, Japan. Although the number of participants was limited, the results provided valuable qualitative insights: the role-playing interaction increased engagement, encouraged perspective-shifting, and facilitated the generation of more varied and reflective questions.

This study investigates the presence and nature of political bias in the translation outputs of three state-of-the-art large language models (LLMs) ‒ ChatGPT-4, Claude 3.5 Sonnet, and Gemini 2.0 ‒ when translating between Hebrew and English in both directions. Focusing on politically sensitive terminology within the context of news reporting, the research examines how each model renders key lexical items such as “terrorist,” “Judea and Samaria,” and “West Bank,” using translation accuracy as a proxy for bias detection. The article presents promising preliminary results of the data-driven analysis of 100 politically diverse news excerpts and a comparative evaluation of translation tendencies across language directions. The findings reveal distinct patterns of ideological framing among the LLMs. ChatGPT-4 exhibited the highest rate of left-oriented bias in Hebrew-to-English translations and a notable degree of right-oriented bias in English-to-Hebrew translations. In contrast, Claude 3.5 Sonnet and Gemini 2.0 demonstrated greater consistency and neutrality, with minimal evidence of politically biased behavior. These results support prior research suggesting that LLMs’ outputs are sensitive not only to the prompt language but also to the underlying training data in specific languages and moderation systems. This study contributes to emerging discussions on AI ethics, multilingual fairness, and the role of post-processing in bias mitigation. The comparative framework presented here offers a foundation for evaluating political bias in multilingual LLM applications, particularly in low-resource language settings.

Large knowledge graphs, such as Wikidata, have immense potential to present all shades of thought and diverse opinions in global public discourse. Understanding and identifying different viewpoints form the basis for research and information systems in various knowledge domains. Hence, this study aims to assess the level of inclusion and representation of multiple viewpoints in Wikidata. This paper proposes a new semi-automatic approach for assessing multiple viewpoint representation within Wikidata, focusing on six inherent mechanisms. The preliminary results reveal that the percentage of items with explicitly presented multiple viewpoints is relatively small compared to the overall number of items in the knowledge base. Wikidata and other large knowledge graphs are widely used as training data and ground truth knowledge bases for AI algorithms and smart decision-making systems. Therefore, building knowledge graphs by ethical principles of inclusion and diversity of viewpoints is a crucial issue.

Generative artificial intelligence (GenAI) tools are gaining popularity in academia, despite concerns about hallucinations and misinformation, which raises questions about their reliability in educational contexts. A key gap lies in the lack of interventions that encourage critical evaluation of GenAI content. Our study addresses this gap by introducing Nudgr, a digital nudging tool designed to promote fact-checking behaviours when using GenAI tools. Nudgr is implemented as a popover alert within GenAI interfaces. It delivers tailored nudge messages, context-relevant keyword suggestions, and one-click verification buttons to direct learners to verify the accuracy of GenAI responses. We evaluated Nudgr with 13 university students, comprising undergraduates and postgraduates across different disciplines. Overall, Nudgr shows promise as a tool that promotes learners’ engagement to verify AI-generated content critically. Its design fosters trust, reduces cognitive effort, and enables faster and more user-friendly fact-checking. It offers a practical response to the challenges posed by misinformation in GenAI use.

To increase the use of research artifacts, such as datasets, it is essential not only to publish them but also to ensure their accessibility. One effective way to enhance accessibility is by revealing relationships between research artifacts, with similarity being a key type of relationship. Identifying the similarities can support the recommendation of research artifacts. A previous study estimated dataset similarity using their metadata. In recent years, methods that estimate sentence similarity through embeddings generated by large language models (LLMs) have achieved strong performance. These findings suggest that LLMs can also be effective for estimating dataset similarity. This paper experimentally verifies the effectiveness of LLMs in estimating similarity between datasets. We implement a similarity estimation method that generates embeddings of dataset metadata using LLMs and computes the cosine similarity between these embeddings. To generate embeddings, we use PromptEOL, which produces high-quality text embeddings by prompting LLMs to capture the meaning of input text in one word. An experiment was conducted to evaluate the estimation performance of the implemented method, and the results demonstrated the effectiveness of the LLM-generated metadata embeddings for estimating dataset similarity

The rapid growth of digital collections has intensified the need for accurate and efficient book classification in digital libraries, yet manual cataloging remains labor-intensive and resource-demanding. Although deep learning approaches achieve strong performance in text classification, their high computational cost and limited interpretability hinder adoption in real-world library environments, particularly in small and medium-sized libraries with constrained resources. This study explores the feasibility of lightweight machine learning (ML) models as practical and resource-efficient methods for automated book genre classification. A curated subset of the Kaggle Books dataset was preprocessed through data cleaning, normalization, and text vectorization, yielding 56,260 records across multiple categories. A set of ML models was evaluated for their effectiveness in automated genre classification. Experimental results show that Logistic Regression outperformed other models, followed by Ridge, LinearSVC, Multinomial Naïve Bayes, and K-Nearest Neighbors, whereas tree-based models demonstrated relatively lower effectiveness and higher computational costs. These findings validate the applicability of linear and probabilistic models for bibliographic categorization, offering a practical entry point for libraries that have not yet explored automation. This research bridges the gap between traditional cataloging and AI-driven knowledge organization by demonstrating that lightweight ML models can serve as effective decision-support tools, particularly for resource-constrained libraries. While full automation remains challenging due to the stringent demands of accuracy and interpretability, incremental adoption of interpretable, resource-efficient models offers a realistic pathway toward Human-in-the-Loop paradigms, mitigating misclassification risks while advancing digital libraries toward more adaptive and intelligent services.

As generative AI systems like ChatGPT become integrated into daily routines, understanding how different users engage with these tools over time is essential for designing future information services. This paper presents an empirical user study examining ChatGPT usage patterns across a diverse set of 38 participants. Using exported chat histories, we analyze over 82,000 prompts spanning up to 21 months per participant. Our findings reveal distinct interaction behaviors tied to age, gender, and time of day. Notably, older users exhibit stronger work-week patterns in usage, while women show higher activity at night. These results highlight the roles generative AI plays in users’ lives, extending beyond productivity into moments of reflection, support, and multitasking.

Digital cultural heritage is increasingly available from memory institutions. However, its integration into formal education remains limited due to a lack of education-specific metadata and weak alignment with curricula. This paper addresses this gap in the Japanese context, where national curricula are centrally defined but rarely linked to digital resources. We present a Linked Open Data (LOD) infrastructure that connects curriculum standards (Course of Study), textbook units, and digital archive content through a reusable, semantically modeled framework. We contribute to implementing a scalable, curriculum-aligned metadata foundation that supports practical educational use. We describe the structure, vocabulary design, and implementation of the infrastructure, and demonstrate its utility through linked applications such as search tools and visualizations. This work lays the groundwork for connecting curricula with cultural resources, enabling more effective discovery, reuse, and integration of archival content in educational settings.

Scientists have always used the studies and research of other researchers to achieve new objectives and perspectives. In particular, employing and operating the measured data in previous studies is so practical. Searching the content of other scientists’ articles is a challenge that researchers have always struggled with. Nowadays, the use of knowledge graphs as a semantic database has helped a lot in saving and retrieving scholarly knowledge. Such technologies are crucial to upgrading traditional search systems to smart knowledge retrieval, which is crucial to getting the most relevant answers for a user query, especially in information and knowledge management. However, in most cases, only the metadata of a paper is searchable, and it is still cumbersome for scientists to have access to the content of the papers. In this paper, we present a novel method of faceted search structured content for comparing and filtering measured data in scholarly knowledge graphs while different units of measurement are used in different studies. This search system proposes applicable units as facets to the user and would dynamically integrate content from further remote knowledge graphs to materialize the scholarly knowledge graph and achieve a higher order of exploration usability on scholarly content, which can be filtered to better satisfy the user’s information needs. The state of the art is that, by using our faceted search system, users can not only search the contents of scientific articles, but also compare and filter heterogeneous data.

In recent years, advances in natural language processing (NLP) have increasingly relied on computational infrastructure, including hardware accelerators, scalable memory systems, software libraries, and framework, and widespread adoption of cloud platforms. However, existing entity recognition methods and scientific knowledge graphs largely overlook these components, instead focusing on research tasks, methods, datasets, and evaluation metrics. To address this gap, we present InfraKG, a large-scale Infrastructure Knowledge Graph that captures and links infrastructure-related entities mentioned in scientific publications. InfraKG is built using a hybrid information extraction framework applied to 85,000 arXiv papers in the computational linguistics domain, combining transformer-based NER models, semantic sentence filtering, and large language models (LLMs). The resulting graph contains 166,728 nodes and 1.5 million relations across seven types, connecting infrastructure entities to scientific publications along with their metadata. InfraKG is the first large-scale resource to systematically represent computational infrastructure in NLP research, enabling advanced queries, trend analysis, and infrastructure-aware literature reviews. We evaluated the proposed framework on 470 manually annotated PDF papers for infrastructure entities, covering a test set of 20,774 sentences. All code and data are publicly available at: code repository .

Documenting cultural heritage has become increasingly critical in the evolving domains of digital libraries, computing, and global information studies. Architecture and its expressions constitute a central component of this endeavor, given their intrinsic connection to human societies and their role as enduring manifestations of cultural heritage. The CIDOC CRM is a well-established and continuously emerging reference model aiming to represent cultural heritage information. This paper establishes a data model for the documentation of vernacular architecture using CIDOC CRM, thereby bridging a critical gap in the representation of culturally embedded architectural practices. In this paper, we discuss the notion of vernacular architecture and the previous work on documenting architecture. Through the classes and properties provided by the CIDOC CRM, we model the formative contexts of vernacular architectural practices, we represent the distinctive constructional, morphological, and typological consistencies inherent in vernacular architecture, and document the interrelations among vernacular stylistic traditions, including their temporal relationships and mutual influences.

This study presents a metadata framework for the digitisation and accessibility of Malaysian historical newspaper collections. It employs a design-based methodology guided by two research questions: (i) What are the metadata requirements and challenges specific to Malaysian historical newspaper archives? and (ii) How can a metadata framework be designed to support effective digitisation, preservation, and user access? The research comprises three essential components: (i) selection of a policy framework; (ii) assessment and evaluation of metadata standards; and (iii) construction of a metadata framework tailored to the requirements of Malaysian historical newspapers. The proposed framework combines standard features with custom fields for cultural and linguistic significance through a comparative analysis of metadata standards, including Dublin Core and its extended DCMI Metadata Terms (ISO/NISO standard), MODS, METS, ALTO, and PREMIS. The prototype, implemented in Omeka, demonstrates a scalable and locally relevant method for cataloguing and conserving Malaysian newspaper information. The framework strengthens efforts to refine metadata practice in heritage digitisation by integrating global standards with national and institutional requirements.

This study addresses the urgent need to preserve the oral heritage of Traditional Malay Midwifery (TMM) through a dual approach: a semantic metadata schema and a domain-specific thematic thesaurus. Using oral history methodology, the framework captures both the content and the cultural depth of TMM practices. The metadata schema, grounded in standards but extended with culturally relevant elements, ensures midwives lived knowledge is accurately represented. The thesaurus, developed from practitioner vocabulary, maps the interconnections of concepts within the traditional knowledge system. Expert evaluations affirmed the model’s relevance, usability, and ethical soundness, emphasizing the importance of culturally responsive archival systems. This research demonstrates that digital preservation of intangible cultural heritage (ICH) must move beyond technical storage to embrace community voice, relational meaning, and ethical design. While developed for TMM, the framework is scalable for other community-based heritage, bridging informal knowledge and structured digital systems.

Hand-drawn anime production generates various intermediate products, such as key drawings, modified key drawings, in-between drawings, and timesheets. These products are paper-based and describe the visual and temporal structure of each cut in a video. However, existing archives generally digitize and store them only as static images, omitting the relationships of timing, layering, and revisions in the animation process. We propose a digital archive system that reconstructs and visualizes these relationships by extracting structural information from timesheets and use it as metadata to link images. We extended the International Image Interoperability Framework (IIIF) manifest to describe frame-by-frame timing, multi-layer composition, and revision histories. Our interactive viewer provides synchronized playback, layer controls, and timeline navigation. In a user experiment involving 20 participants, our system significantly enhanced understanding of motion flow, layer composition, and revision intent compared to a conventional archive system. While participants still found symbolic notations in timesheets difficult to interpret, the proposed system helped them grasp the production process more easily.

The increasing reliance on OCR technologies to digitize documents has enabled large-scale automation but also introduced new challenges for information extraction systems. While state-of-the-art OCR engines perform well under ideal conditions, they remain prone to errors. Traditional OCR evaluation metrics like character and word error rates fail to capture the impact of such errors on downstream tasks, particularly when only semantically critical words are affected. In this paper, we systematically investigate the relationship between OCR quality and extraction accuracy in business documents, with a focus on key field extraction and line item recognition. We introduce a controlled evaluation framework that simulates realistic OCR noise scenarios by selectively injecting errors into clean datasets. Our experiments show that standard OCR metrics poorly reflect the impact of noise on information extraction performance and highlight the need for task-specific OCR evaluation protocols and more resilient pipelines tailored to real-world settings.

This paper presents a practical OCR correction tool developed for digital humanities projects, designed to handle TEI/XML texts generated by various OCR systems including the National Diet Library’s Kotenseki OCR (NDLOCR). By adopting TEI (Text Encoding Initiative) as its data format and leveraging GitHub as the backend infrastructure, the tool enables collaborative editing with version control capabilities across diverse projects. We demonstrate the tool’s effectiveness through real-world application in the Yurenja Catalog Database project. Our approach addresses critical challenges in OCR post-processing while maintaining interoperability with existing TEI-compatible tools and providing a sustainable framework for collaborative text correction in digital humanities.

We investigate how state-of-the-art Large Language Models (LLMs) can unlock new knowledge from the Cyfrowa Biblioteka Druków Ulotnych (CBDU)—a digital collection of early-modern Polish ephemeral prints. Our end-to-end pipeline compares three transcription approaches: pure OCR, LLM-based post-correction, and multimodal models. The resulting transcriptions then serve as input for our two main contributions: the automatic extraction of bibliographic metadata and the generation of expert-style historical commentaries. Experiments show a leading multimodal model excels, reducing transcription CER from 33% to 9%, while achieving high F1-scores for publication place (0.85) and date (0.71), and a 2.31/3 mean score for commentaries. We conclude that large multimodal models can serve as effective “digital archivists”, enriching historical collections with structured metadata and contextual analysis.

National knowledge infrastructures are becoming increasingly vital in higher education as digital transformation accelerates and open-access demands intensify. Malaysia’s National Knowledge Infrastructure (NKI), initiated by the Ministry of Higher Education (MoHE), is a strategic effort to integrate the scholarly resources of 20 public universities into a federated, policy-driven ecosystem. This paper presents a practical case study of the NKI’s early development, documenting the initial planning, design frameworks, and stakeholder strategies that informed its creation. The methodological approach combined policy analysis, environmental scanning, repository audits, international benchmarking, stakeholder consultations, and risk assessment. Planning was guided by the Zachman Framework, which structured activities across four phases—planning, design, implementation, and operation, ensuring coherence between policy objectives and technical design. The FAIR (Findable, Accessible, Interoperable, Reusable) and CARE (Collective Benefit, Authority to Control, Responsibility, Ethics) principles further shaped the initiative by promoting both interoperability and ethical knowledge governance. The feasibility study identified several significant challenges, including diverse repository platforms and metadata standards, licensing restrictions, uneven adoption of open access, and concerns about long-term governance and sustainability. Addressing these issues, lessons learned were synthesized into six strategic pillars: interoperability, licensing and access, open access advocacy, sustainability and governance, equity and inclusion, and monitoring and adaptability. These pillars form the basis of a phased roadmap that connects Malaysia’s immediate next steps with transferable lessons for practitioners designing similar infrastructures elsewhere. By embedding stakeholder participation, policy integration, and inclusive design at every stage, the NKI demonstrates a viable pathway for building national-level infrastructures that are both technically robust and socially responsible. Beyond its national scope, this case study contributes to the broader global discourse on open, federated, and sustainable scholarly infrastructures.

This study examines the publication and citation patterns of data papers—scholarly articles that describe research datasets—across 32 journals indexed in Web of Science. As a cornerstone of open science, data papers enhance transparency, reproducibility, and data reuse. By analyzing articles published in 2023, the study reveals that data papers often receive more citations than traditional research articles in several prominent journals, such as Nucleic Acids Research, Earth System Science Data, and Scientific Data. A new metric, Citation efficiency (CE), is introduced to assess the relative impact of data papers. Journals that employ editorial strategies like special issues tend to achieve higher CE scores, suggesting that visibility and thematic focus contribute to citation performance. While data papers are most common in life and earth sciences, their presence is gradually expanding into the humanities and social sciences. Interestingly, some journals with fewer data papers show disproportionately high citation efficiency, indicating that quality and relevance may outweigh quantity. These findings offer valuable insights for researchers, editors, and policymakers aiming to promote responsible data sharing and improve scholarly communication. By highlighting the impact of data papers, this study supports the advancement of open science and encourages a culture of data-driven research.

This study examines the implementation and impact of the ACM-THU Open transformative agreement at Tsinghua University. Quantitative content analysis of 1,317 ACM publications from 2021 to 2024 by Tsinghua authors reveals that the ACM-THU Open agreement has effectively increased open access publishing, with all articles in 2022 and 2023 being openly accessible. However, challenges remain in author engagement and copyright licensing, as 35% of 2024 publications still used non-Creative Commons (CC) licenses. The study also investigates the relationship between open science practices and research impact. OA articles had significantly higher download counts than non-OA articles, regardless of the presence of open data, code, or materials. Among articles without these open practices, OA status was associated with higher citation counts, but this difference was not significant when open data, code, or materials were present. The findings underscore the impact of various open practices on research visibility and citation advantage. The results suggest that universities should adopt a multifaceted approach to support the implementation of transformative agreements and foster the advancement of open science. This includes establishing stronger mandates for CC-BY licensing, conducting targeted outreach to researchers, and developing integrated information systems to monitor open access compliance and align research practices with institutional open science strategies.

This study aims to analyze the differences in researchers’ data-sharing behaviors to comprehensively measure the acceptance of open scientific data, and to explore the interactions among influencing factors. To elucidate the mechanisms of interaction among influencing factors, a combination of the Big Five personality traits with the COM-B model was proposed, utilizing differential testing methods and structural equation modeling to analyze and validate the questionnaire data. The analysis of the questionnaires showed that different personality traits have varying impacts on data-sharing behaviors. According to COM-B model, the motivation, capability, and opportunity form a positive cycle in promoting data sharing. In the personality traits, conscientiousness and agreeableness significantly and indirectly influence data-sharing intentions.

This study investigated the contribution of dataset reuse to the diversity of research areas using articles that reused datasets. The Framingham Heart Study (FHS) and Atherosclerosis Risk in Communities Study (ARIC), which are widely used in life sciences, were selected as samples. Disease names in the articles that reused these datasets were extracted based on Medical Subject Headings (MeSH) descriptors. The frequencies of disease names were examined from the 1950s for FHS and the 1980s for ARIC. The amount of “cardiovascular diseases” research was decreasing, while that of “pathological conditions, signs, and symptoms” was increasing from the beginning. The Herfindahl-Hirschman Index (HHI) scores, which are an index of diversity, were calculated based on disease names; these scores indicate that research areas have diversified over time. This study found the HHI and MeSH descriptors could be used to measure the contribution of dataset reuse to the diversity of research areas.

Prior knowledge of existing methodologies and resources is essential for a comprehensive understanding of research. In scholarly papers, citations help readers identify prior knowledge. However, research data, such as datasets, lack such mechanisms, making it difficult to trace their origins. In this study, we discuss approaches to acquiring information about the origins of research data, i.e., research data provenance. We focused on scholarly papers as a source for acquiring information about research data provenance, as they often include descriptions of how research data were created. Accordingly, we conducted a preliminary experiment on extracting information about research data provenance from scholarly papers using large language models (LLMs). The results demonstrated the feasibility of extracting information about research data provenance from scholarly papers.

This study examines the usability of a virtual space designed to foster digital resilience and support adolescents affected by cyberbullying. The platform, Cyber Re:New Café, aims to provide a safe, supportive, and empowering space where young people can develop coping strategies, build resilience skills, and engage meaningfully with peers and experts.Usability testing was conducted with ten purposively selected participants, including LGBTQI+ adolescents who had experienced cyberbullying, teacher counselors, and a psychologist. Participants interacted with the platform by completing scenario-based tasks. Data were collected through think-aloud protocols and semi-structured interviews guided by Nielsen’s heuristics, which informed the development of key usability components for youth virtual spaces.Findings indicate that the platform was well-received in terms of accessibility, emotional safety, and user autonomy. However, participants recommended enhancements such as adding shortcut menus for smoother navigation, real-time online status indicators to improve system clarity, and interest-based grouping or designated facilitators to foster safer and more engaging peer interactions. The study synthesized usability components tailored to virtual spaces, which encompass six key dimensions: access, control, feedback, social engagement, emotional resonance, and sustained motivation. This research contributes to HCI by demonstrating how platforms can be designed not only for functional usability but also to promote emotional safety, empowerment, and collective digital resilience among vulnerable youth populations.

Genetic testing has proven effective in health protection and targeted disease prevention, yet its adoption remains limited. Using data from the Health Information National Trends Survey (HINTS 6, N = 3,358), a nationally representative survey in the United States, this study employs hierarchical regression to examine the role of informational and communicative factors in influencing the adoption of genetic testing. The results indicate that individuals’ understanding of medical statistics, anxiety, exposure to genetic testing on the Internet, and interpersonal communication about genetic testing are positively associated with their adoption of genetic testing, whereas mass media exposure to genetic testing is negatively related to their adoption. Additionally, cancer risk perception was not found to be associated with genetic testing adoption. Notably, exposure to content on genetic testing on mass media moderated the effects of anxiety and risk perceptions (perceived susceptibility and severity) on the adoption of genetic testing.

In our digitalized world, people seek information through the internet in everyday contexts, for example, checking bank account balance, comparing insurance policies and buying transport service tickets. Although such online financial services are being used by the public, older people may not access these services due to the grey digital divide. In order to understand their complexities with digital financial services, a tool was created from qualitative data which would enable intermediaries to quantify the factors that could contribute to the challenges faced by older people. This paper discusses how the tool was created and how it could be leveraged to facilitate older people in accessing online financial services. This tool would also assist in identifying the prevalent and diverse challenges encountered by older people from different cohorts, for example, specific ethnic minority groups. Common patterns could emerge from the data acquired through this tool that could aid in creating interventions which would reduce digital exclusion.

Existing research underscores the critical role of the information environment (IE) in shaping students’ transition to university, yet the specific challenges they face in navigating this landscape remain underexplored, particularly in ways that can inform targeted support. This study addresses that gap by examining the experiences of 34 incoming first-year students from Generation Z (Gen Z). Drawing on qualitative interviews, this panel study explores how the shift from a “kind” pre-tertiary IE—defined by clarity, guidance, and reliability—to a “wicked” IE marked by fragmentation and ambiguity creates new difficulties for students in transition. Findings highlight three key areas of change: students’ evolving role from passive recipients to active information seekers, the move from centralised and authoritative sources to decentralised and less structured ones, and a perceived decline in information quality due to the absence of traditional gatekeepers. These insights situate Gen Z’s information practices within a rapidly evolving information landscape during the transition process, underscoring the need for higher education institutions to strengthen informational scaffolding through integrated official platforms, more comprehensive and transparent content, and targeted information literacy training.

The Japanese national curriculum increasingly emphasises inquiry-based learning (IBL), requiring school librarians to support this process by providing curated book lists. While librarians often rely on existing examples, the diverse nature of IBL makes relevant cases scarce. To address this gap, we present an online social platform, BookReach, designed for Japanese school librarians to create, share, and adapt book lists for IBL classes. This demonstration paper introduces the platform’s architecture, which integrates a unit-based curation interface with community features for sharing and adapting ‘practice cases’. A usability study with 18 participants yielded an average System Usability Scale (SUS) score of 62.36. While indicating acceptable usability, this result, when combined with qualitative feedback, highlights the critical need for open metadata on book difficulty. We will extend the platform for more social features and support longitudinal studies of real-world curation practices.

Freedom of Information (FOI) initiatives are vital for democratic transparency, but in the Philippines, the implementation of public FOI portals has raised critical privacy concerns. This paper examines the Philippine eFOI portal, a government-run platform that publishes FOI requests and agency responses online, often revealing users’ personal information. Using a dataset of over 235,000 FOI requests between 2016 and 2024, along with Presidio, an open-source personal information analyzer, the study identifies widespread inclusion of personal and sensitive personal information despite existing privacy advisories. Analysis reveals that 44% of requests contained personal information, and 0.66% included sensitive identifiers, with agencies like OWWA and SSS having disproportionately high exposure rates. This trend is particularly acute during crisis periods like the COVID-19 pandemic, when citizens used FOI as a substitute for service portals. Results highlight systemic design flaws in the portal, user misunderstanding of FOI scope, and the absence of adequate redaction protocols. Drawing on international examples from the US, UK, and Australia, the paper argues for integrating automated de-identification tools with community oversight to promote responsible transparency. It concludes by advocating for a comprehensive FOI law in the Philippines that ensures legal safeguards, institutional accountability, and privacy protection without undermining openness.

Balancing the value release of data assets with privacy and security poses significant challenges in the era of data elementization. This paper constructs a privacy risk assessment model integrating Differential Privacy (DP) and Statistical Disclosure Control (SDC), aiming to provide a quantitative basis for setting privacy protection intensity and selecting strategies for data assets. The study first elucidates the core principles of DP and SDC and their risk quantification foundations, then proposes a privacy risk calculation framework. Using hospital electronic medical record data assets as an empirical subject, six core attributes and health risk scores are selected as variables. Four access control models simulate physician access behaviors, and their privacy protection efficacy is evaluated under DP and SDC frameworks. Results show that under the DP framework, Attribute-Based Access Control (ABAC) best satisfies differential privacy conditions. Under the SDC framework, Mandatory Access Control (MAC) exhibits the lowest privacy leakage risk, while Role-Based Access Control (RBAC) shows the highest. Furthermore, the contribution of key variables to risk depends on their characteristics and mechanisms. This study provides actionable references for data asset management institutions in designing privacy protection schemes.

This study conducted a comprehensive review to elucidate research trends in data publication and sharing. 6,773 articles on data publication or sharing published between 2022 and 2024 were analyzed and assigned categories related to topics on data publication. The most common topic was “Legal systems, policies, and governance.” This category included articles discussing legal issues related to data and the impact of legal systems on data sharing. The second most common topic was “Status and impact of data sharing,” including articles on challenges in data sharing and researchers’ motivations for sharing data. In the medical field, many studies focused on “Privacy protection” and “Factors and consent for data sharing.” In contrast, research belonging to categories “Data reuse” and “Data quality” remains scarce. This review revealed that current research focuses on the infrastructure required for data sharing and the analysis of existing conditions and challenges before data publication. Meanwhile, studies addressing the post-publication phase are still limited. A systematic review focusing on post-publication aspects, such as data reuse and data quality, will be conducted.

Generative artificial intelligence (GenAI) is increasingly playing a pivotal role in cultural heritage activation and engagement which has not been adequately understood. This study explores the role of GenAI in activating and engaging cultural heritage with a focus on how GenAI can benefit restoration, storytelling, and interaction with the cultural heritage at the Dunhuang Mogao Cave Exhibition using a direct observation method. It shows that GenAI technologies such as image-to-video, image rectification and inpainting, text-to-speech, and temporal sequence generation can boost dynamic activation and interactive narrative in the exhibition. This leads to the conclusion that GenAI is a transformative tool in the activation and engagement of cultural heritage.

Documentary heritage constitutes a crucial witness to the Silk Road, underpinning the cultural foundation of the Belt and Road Initiative. Despite mature preservation mechanisms, there exists an imbalance between micro-level semantic analysis and macro-level cultural interpretation. To bridge this gap, the study proposes an underlying logic for decoding and translating cultural genes in China’s overland Silk Road heritage, and develops a three-stage framework based on semantic expression, associative aggregation, and scenario reconstruction. Using Zhang Qian’s missions to the Western Regions as a case, the research constructs a multi-layered semantic system from 273 data objects, encompassing resource, knowledge, and service layers. The results demonstrate how cultural genes embedded in documentary heritage can be systematically identified, semantically linked, and digitally activated, offering a novel perspective for understanding the cultural evolution of the Silk Road.

Generating high-quality descriptions for web archives remains a time-consuming bottleneck in digital preservation workflows. This paper explores the use of large language models (LLMs) to automate this task, focusing on fine-tuning the Qwen3-8B model on a curated corpus of human-written summaries. The resulting system produces semantically accurate and context-aware descriptions of HTML records extracted from WARC files. We integrate the model into a full-stack pipeline that handles ingestion, parsing, and AI-driven analysis. Performance is evaluated using BERTScore and MoverScore, alongside manual assessments by archivists, librarians, and archival science students. Results show low Composite Edit Rates (CER) and high user satisfaction, validating both the reliability and utility of LLMs for metadata generation. These findings highlight the potential of AI to enhance scalability, consistency, and quality in archival description practices.

Automatic subject indexing is increasingly implemented using Large Language Models, whose performance, however fluctuates sharply across disciplines. A two-dimensional framework—breadth (surface term coverage) and depth—is introduced to to quantify the structural quality of disciplinary controlled vocabularies. Based on the German GND authority file across ten disciplines, several indicators are computed, Spearman correlation analysis reveals that higher breadth is significantly improves indexing accuracy, while excessive depth correlates negatively with performance. These findings confirm that performance hinges not on maximizing either dimension, but on achieving a breadth–depth balance. This work provides the empirical evidence that breadth–depth balance, rather than maximizing either alone, governs cross-disciplinary indexing performance, offering actionable guidance for building equitable, discipline-sensitive, LLM-compatible knowledge bases.

This study reviews Digital Humanities (DH) research to compare knowledge organization (KO) approaches between traditional DH and emerging Artificial Intelligence Humanities (AI Humanities). Using content analysis, we systematically examine how KO is applied in traditional DH and AI humanities. Traditional DH projects follow an “extraction-representation-storage-application” knowledge organization paradigm, focusing on improving the discoverability, accessibility, and comprehensibility of humanities knowledge. Generative AI technology widely influences and transforms the way of knowledge organization, shifting from a high technical threshold and professional knowledge requirements to simple natural language guidance, from rule-based frameworks to automatic processing with LLMs, from domain expert-led practices to intelligent user interactions, and from knowledge browsing to knowledge generation. Based on these shifts, the study proposes future directions for knowledge organization in AI humanities, providing insights for further research.

This paper examines the often-invisible aspects of labor of the staff and volunteers at the Bantayog ng mga Bayani, a memorial institution in the Philippines dedicated to honoring the martyrs and heroes who fought the Marcos dictatorship. Using archival ethnography informed by community archives and feminist ethics of care, it highlights how archival work at the Bantayog ng mga Bayani is sustained through emotional, relational, and political forms of care. It argues that making these aspects visible is an ethical and political imperative, especially in under-resourced contexts deeply intertwined with trauma and resistance. In centering the lived experiences of archival workers, the paper contributes to broader discussions of care, sustainability, and justice in archives.

This study examines the household information practices of thirteen Filipino middle-income households residing in the National Capital Region (NCR) of the Philippines, drawing on Dervin’s sense-making framework and Kalms’ Theory for the Emergence of Household Information Practices. Through semi-structured interviews, the research investigates the roles, relationships, and dynamics of how Filipino households navigate the responsibilities and challenges of handling records in the household. Results show that the information practices are shaped by traditional family structures. Parents, especially mothers, act as primary recordkeepers, while older children are expected to fulfill support obligations to their younger siblings and sometimes even to their parents. These practices reflect broader expectations rooted in age, gender, and familial hierarchy. The study of their perspectives also reveals concerns on accessibility and change in practices arising from additions to the household and life transitions such as education, employment, and coming of age. Overall, the findings contribute to a deeper understanding of how Filipino families negotiate the organization, delegation, and transfer of information-related responsibilities in the domestic sphere.

Archivists must navigate increasingly complex moral dilemmas, yet little empirical work describes how Southeast Asian programs cultivate ethical competence. Addressing this gap, the present paper asks: how, and to what extent, ethics are taught in archival education? A qualitative comparative case study analyzed policy documents, syllabi, and assessment rubrics from two flagship institutions: the Master in Archives and Records Management (MARM) at the University of the Philippines Diliman and the archival-studies programs at the University of Social Sciences and Humanities, Vietnam National University Ho Chi Minh City. Findings reveal contrasting curricular philosophies. These models mirror national conditions: the Philippines compensates for limited legislation through curricular creativity, while Vietnam secures uniform conduct via legislation. The paper suggests that a hybrid approach marrying reflexive critique with regulatory literacy would better prepare Southeast Asian archivists for culturally plural, digital environments, as the cases show. By documenting current practice, it strengthens global debates on ethics pedagogy and offers region-specific guidance for curriculum designers.

The research aims to analyse archive collections and examine the process of archives management of the Debsirin Alumni Association by implementing the Records Continuum Model. The above objective aims to provide a comprehensive overview of the issues and archives management process of the association, leading to the creation of a practical approach for applying the RCM in their practice for managing records and archives systems that support the operational needs, the preservation of institutional memory within school communities, institutional accountability, and school reputation in the digital era.