Abstract This study evaluated the performance of Turnitin, a tool widely used across colleges and universities for detecting AI-generated text. Our sample of human-written text comprised 100 introduction sections from psychology journal articles published in 2016. Next, we generated new versions of the texts using ChatGPT (GPT-4o and GPT-5). Using Turnitin’s AI score for each piece of writing, we found a false positive rate (incorrectly classifying human-written text as AI-generated) of only 1 %. In contrast, Turnitin’s false negative rate (incorrectly classifying AI-generated text as human-written) was substantially higher and depended on the model used to generate the text. Specifically, the latest version of ChatGPT produced text more likely to be misclassified as human-written than the previous version. These findings highlight important limitations in the reliability of automated AI-detection tools, particularly with respect to false negative rates and model-dependent performance. As such, AI-detection scores should be interpreted cautiously, especially when used to infer the presence or absence of AI-generated text in academic writing.
This study assessed the availability, accessibility, and utilization of Learning Resource Centers (LRCs) in rural Philippines to guide policies and community initiatives promoting educational inclusivity. Using a descriptive–correlational design, data were collected from public library users via a validated, pilot-tested researcher-developed questionnaire. Analyses included frequency counts, percentages, means, standard deviations, t-test, ANOVA, and Pearson’s r using SPSS 17.0. Most respondents were female, aged 18–28, and students. Findings showed moderate availability and accessibility of resources and services but low utilization. No significant sex-based differences were found in availability or utilization, although females perceived greater accessibility. ANOVA revealed significant differences across age groups and user types, suggesting demographic factors influence engagement with LRCs. Correlation analysis indicated strong positive relationships among availability, accessibility, and utilization, implying that improvements in one area enhance the others. The study underscores the need to strengthen rural LRCs through enhanced infrastructure, updated collections, technological integration, and user-centered programs to promote equitable access and effective utilization of learning resources.
Graduate Information Science education in the Asia–Pacific (AP) region has diversified in response to digital transformation and the globalization of knowledge work. This study investigates how curriculum characteristics and curriculum identities of graduate Information Science programs relate to international collaboration needs across the AP iSchools. A sequential explanatory mixed-methods design was adopted. Phase 1 involved qualitative documentary analysis of 37 Master’s and Doctoral curricula to identify structural characteristics and derive six curriculum identity clusters. Phase 2 consisted of a quantitative survey of 55 participants, 26 faculty/program leaders and 29 graduate students to assess collaboration needs across five dimensions. Findings indicate a shared disciplinary foundation across programs but variation in specialization orientation, forming six identities: classical LIS, archives and records, information management, informatics/data analytics, community/social development LIS, and digital governance. Stakeholders showed strong support for collaboration, though priorities differed: faculty emphasized research productivity and institutional visibility, while students prioritized mobility and employability. Integrated results highlight that collaboration initiatives are most effective when aligned with curriculum identity and responsive to diverse stakeholder needs.
The rapid advancement of artificial intelligence (AI) is reshaping research practices and scholarly communication within higher education institutions and academic libraries. This systematic review examines the integration of AI tools in research and scholarly communication, focusing on their benefits, challenges, and the evolving role of librarians in this digital transformation. Guided by the preferred reporting items for systematic reviews and meta-analyses (PRISMA) framework, the study retrieved literature from EBSCOhost (Library and Information Science Source and Library and Information Science & Technology Abstracts), ProQuest, and Google Scholar, covering publications from 2020 to 2025. Of the 226 articles initially identified, 24 met the inclusion criteria and were subjected to thematic analysis. The findings reveal that AI technologies significantly enhance research efficiency, improve information retrieval, support data analysis, and streamline scholarly communication workflows. At the same time, the review identifies critical challenges, including ethical concerns, data privacy risks, algorithmic bias, and insufficient AI literacy among information professionals. Academic librarians emerge as key facilitators in the responsible adoption of AI, requiring ongoing professional development to navigate these technological shifts effectively. The study concludes with recommendations for the ethical and sustainable integration of AI tools in academic research environments and identifies gaps in the literature that require further scholarly attention to ensure equitable and responsible AI implementation in scholarly communication systems.
This study investigates the information literacy skills of Senior High School students in Ghana, concentrating on their capacity to access, assess, and ethically utilise information in the digital age. Employing the cross-sectional correlational design and adopting the quantitative approach by using the Information and Communication Assessment Instrument (ICAI), data were gathered from 800 students across nine schools, evaluating skills across ten areas. Findings show a moderate average information competency score of 178.17 (SD = 24.61) out of 280, with 73 % of students classified as having medium competency, 12.8 % high, and 14.3 % low. Students performed well in locating and retrieving information (M = 18.53, SD = 4.71) but showed gaps in using information technologies (M = 16.65, SD = 4.76) and understanding ethics and legality (M = 16.93, SD = 4.40). These findings align with global trends, showing strengths in basic information retrieval but persistent weaknesses in critical evaluation, synthesis, and technological skills. In Ghana, systemic challenges, including inadequate ICT infrastructure, insufficient teacher training, and limited curriculum integration, worsen these issues. The study recommends stronger collaboration between librarians and teachers, curriculum reforms to incorporate information literacy, better technological infrastructure, and targeted professional development to address these gaps and prepare students for higher education and knowledge-driven careers.
The global movement toward Open Science presents an unprecedented opportunity to bridge knowledge gaps and promote inclusivity in research and development. Promoting Open Science and open access to knowledge can significantly contribute to the collective building of national economies and lead to societal progress. This study aimed to investigate the possibility of building an inclusive knowledge society in Uganda through embracing the Open Science best practices at the GESIS – Leibniz Institute for the Social Sciences in Germany. An institutional survey was conducted by interviewing eight key informants and experts in Open Science at GESIS. The Open Science practices explored include: the regulatory framework, Open Science infrastructure, funding, education and training, data quality & security, and stakeholder engagement. The paper further assesses the current challenges and proposes strategic interventions tailored to Uganda’s socio-economic and technological landscape. Emphasis is placed on equity in access to scientific knowledge, infrastructure development, policy, and capacity building. The paper also highlights recommendations for a multi-stakeholder approach to sustainably integrate Open Science into Uganda and other developing economies’ national development agendas.
The integration of artificial intelligence (AI) into automatic recommendation systems for improving university guidance represents a major advancement in educational support services. Effective guidance helps students navigate their academic journey, make informed decisions about courses and career paths, and achieve their personal and professional goals. AI-powered systems offer personalized recommendations that can transform how students plan their future by delivering precise, tailored advice. Following PRISMA guidelines, we conducted a systematic literature review, analyzing 28 relevant studies from a pool of 451 articles published between 2013 and 2024. This review aimed to explore the influence of recommendation systems on student guidance, the development methods employed, the types of data used, and the evaluation metrics applied. Our findings highlight that machine learning and hybrid approaches improve the understanding of student needs and behaviors, enabling more accurate and relevant recommendations. Furthermore, these systems can contribute to lower dropout rates and enhanced student success. The most commonly used data types include academic data, personal and preference data, demographic data, family and social background data, and skills and competency data.
Getting published in globally indexed journals is still a major hurdle, especially for those scholars working in the still-developing academic systems. This study seeks to understand these barriers at the underlying level by surveying 120 researchers from Hue University and Times Higher Education (THE) and Quacquarelli Symonds (QS)-ranked universities. Deploying quantitative analysis and machine learning (ML) techniques, such as Random Forest (RF), Decision Tree (DT), and Gradient Boosting (GB), showed that financial burdens, institutional constraints, and limited collaboration opportunities were the three biggest obstacles. RF was found to have the highest classification accuracy among all models in predicting role and discipline barrier levels. Main recommendations highlight the calls for targeted interventions: subsidy for publication costs and enforce transparent article processing charges (APCs) waiver practices (Q1. High Publication Fees) Encourage transnational collaboration through exchange programs and conference sponsorship (Q7. Language/Cultural barriers) Role-sensitive funding models to address inequities within and between disciplines and academic ranks (Q12. Lack of Metadata Knowledge) Improving access to centralized data repositories and offering more technical/analytical training (Q6. Insufficient Funding, Q10. Lack of Access to Trending Data), as well as improving mentorship frameworks for early-career researchers (Q5. Lack of International Networks, Q12. Lack of Metadata Knowledge), are key structural fixes to inequitable disadvantages. Together, these strategies lower barriers while encouraging larger and more diverse participation in global higher education. The main purpose of the present study is to identify the problems encountered by the researchers in publishing their articles in the Web of Science (WoS) and Scopus-indexed journals. Above all, this study is to lay out how the major barriers, their intermingling and working opposition to one another, and what institutional action is needed to realize improved rates of successful publication. To do this, a combination of quantitative analysis and ML is used to help inform and tell a more robust narrative around the effect of the researchers’ published work in high-impact publications. These quantitative methods and ML techniques are developed based on an empirical survey of 120 scholars from Hue University and globally ranked institutions. To collect data, a structured questionnaire on 12 factors affecting academic performance was used. Descriptive statistics, correlation matrix, ANOVA, and Structural Equation Modeling (SEM) were used to analyze the data. At the same time, RF, DT, and GB models were trained to classify and predict barrier levels. The biggest obstacles noted were the high cost of publishing in science, technology, engineering, and math (STEM) fields, the lack of global coordination and collaboration, and unequal access to resources. The RF model performed best in classification accuracy compared to the DT and GB models. Recommendations included subsidizing conference fees, developing mentorship programs, expanding access to industry data, and role-based funding. Together, these efforts would eliminate existing publication inequities. Novelty: This study was groundbreaking in ML with quantitative analysis methods of combination to proactively identify and address these publication barriers. These findings provide additional ammunition for our policymakers, research administrators, and scholars, who all need to be armed with enough knowledge to work toward an equitable scholarly publishing environment.
This study aimed to examine chatbot use, perceived concerns, effects, and ethics among university students. The study adopted a purposive sampling technique to select participants. The data for this study were collected from university students from India, the UAE and Jordan. A structured questionnaire was developed from published literature. The study instrument was planned to collect data on respondents’ familiarity and usage, measuring the familiarity and unfamiliarity of different chatbots such as Bard AI, CoPilot, etc., and the frequency of use is measured in terms of using the chatbot regularly, rarely, and never. While attitude, effects of chatbots, and ethical aspects are measured on 5-point Likert scale with 1 signifying strongly agree and 5 signifying strongly disagree. Data were collected electronically through email distribution of the designed Google form survey. A total of 444 participants in this study. The six hypotheses were tested using the independent sample t-test and the ANOVA test. The results of the study present significant differences in the use of chatbots across India, the UAE, and Jordan. The study showed differences in gender and academic level in chatbot use. The analysis exposed that science students demonstrated significantly higher levels of chatbot use and concern compared to non-science students. The outcomes of the study provide valuable insights into the integration of chatbots in academic institutions.
Recommendation systems frequently exhibit two fundamental types of bias: user-side bias, where recommendations discriminate based on protected attributes such as gender, and item-side bias, where popular items dominate exposure irrespective of their actual relevance or quality. To address both simultaneously, we propose a dual-adversarial framework with two discriminators: a user-gender discriminator designed to mitigate gender bias by removing sensitive information from user embeddings, and an item-popularity discriminator aimed at reducing popularity bias among items. While adversarial learning encourages the model to generate bias-free embeddings by fooling discriminators, residual sensitive information may persist due to imperfect discriminator inference and distributional discrepancies. To further reduce this residual bias and enforce balanced, fair recommendations, we enhance the dual adversarial learning framework with two explicit fairness constraints that operate on the learned debiased user and item embeddings: an individual user fairness constraint that guarantees equitable recommendations for users with similar preferences, and a popularity regularization term that balances exposure across popular and less popular item groups. Experiments on three real-world datasets (MovieLens 1M, MovieLens 100K, and Book-Crossing) demonstrate significant gender predictability drops and popularity skew decreases, while maintaining competitive recommendation quality.
Both online and in-person social networks can enhance peer support for people with disabilities when they are designed in line with current accessibility standards and validated through real-world use. Drawing on an integrative review and brief case studies, we identify seven key barriers – stigma, inaccessible interfaces, the digital divide, biased or opaque moderation, privacy and security concerns, assistive-technology gaps, and fragmented communities – and link them to practical, standards-based solutions. We outline release-blocking checks for essential social tasks aligned with WCAG 2.2 and EN 301 549, including contrast, keyboard operability, visible focus, error handling and prevention, non-drag alternatives, minimum target sizes, and consistent help features. In addition, we propose actionable performance metrics such as first-post and join rates, 30- and 90-day retention, task success and completion time, caption and alt-text coverage, keyboard-only pass rates, and moderation error rates. Case vignettes further demonstrate how concepts from network theory – such as weak ties, brokerage, and closure – can inform design decisions and measurable outcomes. The study concludes with ethical safeguards and a testing framework that emphasizes low-bandwidth delivery, interoperability with assistive technologies, and audited, explainable moderation to support the scalable delivery of inclusive peer.
Students worldwide use digital technologies such as mobile phones, smartphones, computers, tablets, social media, the internet, and digital televisions. This scoping review aims to map the literature on the types of digital technologies and the purposes of using digital technologies by secondary school students. The Scopus, Web of Science, and ProQuest Library Science databases were searched to identify and retrieve literature from 2000 to 2024 after a search strategy was developed. The study is limited to secondary school students. This scoping review followed the PRISMA-ScR guidelines. Of the 2,246 records, only 35 studies that met the inclusion criteria were considered for this review. This review provides insight into the types and purposes of the use of digital technologies by secondary school students. Studies have revealed that digital technologies are used for academic purposes and thus promote reading habits. Moreover, studies have indicated that e-books, iPads, e-readers, audiobooks, social media, and smartphones are widely used for reading and information access. However, few studies have shown that the overuse of digital technologies for entertainment leads to a decline in reading habits. Further research is recommended to explore how digital technologies could be used to promote sustained and meaningful reading practices among secondary school students.
This study explores the present status of IoT adoption and implementation in Information Resource Centres (IRCs) of Institutions of National Importance (INIs) in India. It aims to identify the extent of integration, commonly adopted technologies, and challenges faced, proposing insights for strategic implementation. A quantitative survey was conducted among IRC administrators from 18 selected INIs in South India, including Indian Institutes of Technology (IITs = 5), Indian Institutes of Management (IIMs = 4), Indian Institute of Science (IISc = 1), National Institutes of Technology (NITs = 5), and All India Institutes of Medical Sciences (AIIMS = 3). Data were collected via physical questionnaires and online Google Forms. Non-parametric statistical tests were used to analyze demographic patterns, the present status of IoT adoption and challenges across different institution types. The response rate was 88.88 % and reported some level of IoT adoption, reflecting positive institutional attitudes. Basic IoT services such as real-time notifications, QR codes, and Radio Frequency Identification (RFID) tracking display high and uniform implementation across IRCs with high mean values (≥4.00). In contrast, advanced applications like virtual/augmented realities, smart infrastructure, and mobile-based tools remain underutilized (≤M = 2.20), primarily due to cost, technical difficulty, and personnel limitations. The study also shows no statistically significant difference in IoT adoption levels across the various types of INIs (p > 0.05), representing a consistent pattern in digital transformation efforts. Key challenges identified include financial constraints, data security concerns, limited strategic planning, and lack of trained personnel. The study focuses on selected INIs in South India and provides an inclusive snapshot of the IoT implementation landscape in top-tier academic libraries and suggests effective strategies for scaling up adoption efforts. This is an original and first study exclusively on different categories of premier academic institutions allied with the given status of INIs, to investigate the current status of IoT adoption and implementation in IRCs. This empirical indication contributes to guide policy, planning, and future research on smart library transformation.
This study presents a comprehensive bibliometric analysis of data science research in Ecuador from 1985 to 2023, aiming to trace its academic development and highlight real-world applications. Rather than evaluating the direct integration of data science into national infrastructure, the study focuses on how the Ecuadorian scientific community has contributed to this field’s evolution. The analysis is divided into two stages: moderate growth (1985–2015) and exponential expansion (2016–2023), with a strong correlation between scientific output and legislative reforms such as the Organic Law of Higher Education and the Prometheus Project. Additionally, selected case studies illustrate how data science has been applied in healthcare, education, and business through technologies such as IBM Watson, Microsoft Azure, and machine learning models. This dual approach – bibliometric and applicative – sheds light on Ecuador’s trajectory in scientific production and technological adoption, offering insight into the nation’s research landscape and future innovation potential.
The rapid expansion of scholarly publishing has amplified the long-standing challenge of author name ambiguity in academic databases. This issue, manifesting as homonymy and synonymy, undermines the accuracy of bibliometric analyses, author-level metrics, and research evaluation systems. Author Name Disambiguation (AND) has thus emerged as a critical focus area in digital scholarship, with evolving strategies ranging from supervised machine learning and graph-based models to the adoption of persistent digital identifiers like ORCID. Despite notable advancements, significant challenges remain – particularly in linguistically diverse and underrepresented regions – where metadata inconsistencies, transliteration issues, and limited ORCID adoption exacerbate disambiguation errors. This study presents a comprehensive bibliometric analysis of 2,004 publications on AND from 2005 to 2024, sourced from the Scopus database. Using tools such as Biblioshiny and VOSviewer, the analysis identifies publication trends, leading authors and institutions, core sources, co-authorship networks, and thematic evolution in the field. Findings highlight increasing international collaboration, the dominance of computer science-driven methodologies, and the critical role of metadata quality and institutional frameworks. The study concludes with recommendations for inclusive, multilingual, and interoperable disambiguation systems, advocating for cross-disciplinary collaboration to ensure equitable author identification in global scholarly communication.
This study investigates student perceptions of artificial intelligence (AI) adoption in university libraries, focusing on their understanding of AI’s benefits and ethical considerations essential for broader institutional implementation. A quantitative survey was conducted using structured questionnaires distributed among 400 students from Kurukshetra University, Shri Krishna Ayush University, NIT, and NID, with 377 valid responses analyzed using SPSS. Statistical tools such as ANOVA and percentage analysis were employed to assess the influence of demographic and institutional factors on students’ perceptions and expectations. The findings reveal that students view AI positively, recognizing its role in improving access to resources, personalized recommendations, and operational efficiency. However, concerns persist regarding data privacy, reduced human interaction, and job displacement. Notably, students with greater exposure to AI showed higher acceptance, while others remained cautious. The study’s limitations include reliance on data from four institutions and self-reported responses. Nonetheless, the results provide valuable insights for academic libraries across sectors. Practical implications suggest the need for privacy safeguards, transparent AI systems, and targeted training to build trust and enhance adoption. This research contributes to academic discourse by aligning AI implementation with student needs and ethical standards, offering institutional strategies for effective AI integration in library services.
This study presents a comprehensive bibliometric analysis of intangible cultural heritage (ICH) literature published between 2005 and 2024, using data retrieved from the Scopus database. A total of 4,649 documents were analyzed to uncover publication trends, key contributors, influential sources, thematic evolution, and international research collaboration. The analysis reveals a consistent annual growth rate of 10.15 %, with a notable surge in scholarly interest in recent years. Core publication sources include journals focused on folklore, ethnobotany, and heritage studies. Author productivity follows Lotka’s Law, with a small group of prolific contributors shaping the discourse. The keyword co-occurrence network highlights conceptual anchors such as “indigenous knowledge,” “traditional knowledge,” “folklore,” and “UNESCO,” while also revealing intersections with emerging themes like sustainability, climate change, and traditional medicine. Country-level co-authorship analysis shows strong research output and collaboration from the United States, South Africa, the United Kingdom, and a growing presence from countries in Asia and Africa. This study provides a structured knowledge map of ICH research, identifies foundational works, and offers insights for future scholarly inquiry and policy development in cultural heritage preservation.
This study examines the legal framework governing public libraries in India to assess the integration of green library principles into existing legislation. The research analyzes the Library Acts of 19 Indian states, focusing on the presence or absence of terminology related to “green libraries,” “eco-friendly” practices, and “sustainability.” The analysis reveals a significant policy gap, as none of the Acts, regardless of their date of enactment, include language promoting environmental responsibility in library operations. While individual libraries may be undertaking green initiatives, a supportive legal framework is essential for widespread adoption and consistent implementation of sustainable practices. The study recommends that these guidelines encompass aspects such as sustainable building design, resource conservation, waste management and environmental education programs for library patrons.
In a metric-driven publishing landscape, local journals often struggle for visibility and recognition, despite their potential to inform national policy. This study investigates the policy relevance of Australian journals using citation data from the Overton database. By comparing citations in policy documents to Australian-authored articles published in local versus international journals between 2010 and 2022, we find that local journals are more likely to be cited in policy documents, especially those related to Australia. However, the average number of citations per article is lower for local journals, which probably reflects their niche focus and limited international uptake. These findings illustrate the vital but undervalued role national journals play in supporting local scholarship and addressing context-specific issues. We argue that greater institutional support is needed to sustain national publishing ecosystems and prevent the marginalisation of research with local relevance.
Misinformation about agricultural government schemes often impedes farmers from fully utilizing available initiatives. This study examines the role of information literacy in mitigating misinformation and enhancing farmers’ access to government support. The study employed a mixed-methods approach with a simple random sampling technique to select 200 farmers from four villages in West Jaintia Hills, Meghalaya. Data collection included surveys, interviews, and focus group discussions to assess information literacy and misinformation’s impact on trust. The findings reveal that while most farmers are aware of agricultural schemes, they predominantly rely on informal sources such as peers and family, with 36.5% citing them as their primary source, highlighting the role of social networks in information dissemination. The key challenges include low digital literacy, limited access to verified sources, and social influences affecting information trust. Statistical analysis indicates negligible correlations between education or experience and information literacy, though strong correlations exist among literacy skills. The study also contributes original insights into the relationship between information literacy and misinformation, emphasizing the need for targeted literacy training, improved government communication, and community-driven interventions to enhance farmers’ decision-making and participation in agricultural schemes.
