The exponential growth of user-generated video content necessitates efficient summarization systems for improved accessibility, retrieval, and analysis. This study presents and benchmarks a multimodal video summarization framework that classifies segments as informative or non-informative using audio, visual, and fused features. Sixty hours of annotated video across ten diverse categories were analyzed. Audio features were extracted with pyAudioAnalysis, while visual features (colour histograms, optical flow, object detection, facial recognition) were derived using OpenCV. Six supervised classifiers—Naive Bayes, K-Nearest Neighbors, Logistic Regression, Decision Tree, Random Forest, and XGBoost—were evaluated, with hyperparameters optimized via grid search. Temporal coherence was enhanced using median filtering. Random Forest achieved the best performance, with 74% AUC on fused features and a 3% F1-score gain after post-processing. Spectral flux, grayscale histograms, and optical flow emerged as key discriminative features. The best model was deployed as a practical web service using TensorFlow and Flask, integrating informative segment detection with subtitle generation via beam search to ensure coherence and coverage. System-level evaluation demonstrated low latency and efficient resource utilization under load. Overall, the results confirm the strength of multimodal fusion and ensemble learning for video summarization and highlight their potential for real-world applications in surveillance, digital archiving, and online education.

The development of technology has influenced changes in agricultural production. Farmers are increasingly using modern devices and machinery that provide valuable information, and to manage this information effectively, it is necessary to use specialized applications. This research aims to evaluate various applications and determine which one is most suitable for small- and medium-sized farmers to adopt in precision agriculture. This research employed expert decision-making to determine the importance of criteria and evaluate applications using linguistic values. Due to the presence of uncertainty in decision-making, an interval type-2 fuzzy (IT2F) set was used, which addresses this problem through the support of a membership function. This approach allows for the display of uncertainty and imprecision using an interval rather than a single exact value. This enables a more flexible and realistic representation of ratings, leading to more confident decision-making. These membership functions are formed in such a way that there is symmetry around the central linguistic value. To use this approach, the SiWeC (simple weight calculation) and CORASO (compromise ranking from alternative solutions) methods were adapted. The results of the IT2F SiWeC method revealed that the most important criteria for experts are data accuracy, efficiency, and simplicity. The results of the IT2F CORASO method displayed that the A3 application delivers the best results, confirmed by additional analyses. This research has indicated that digital tools, in the form of applications, can be effectively used in small- and medium-scale precision agriculture production.

The progress of Bosnia and Herzegovina in the process of EU integrations can be described as slow and difficult. Many reasons contribute to such development, starting with political instability. However, even in the case of absence of political difficulties, certain features of the constitutional and legal system of the country make decision-making at state level and conducting necessary reforms slow and ineffective. Bosnia and Herzegovina’s constitution is a part of a Dayton Peace Agreement, which was a result of long peace negotiations to end one of the bloodiest conflicts in the 1990’s. It achieved in stopping the conflict, but inserted many features which make reforms difficult, such as complicated decision-making process, fragmented state apparatus and division of competences between different levels of government. One of the features of the constitutional system is the set of group rights belonging to the so-called constituent peoples. The principle foresees certain procedural rules related to the parity in appointment of officials as well as necessary quotas and veto powers in the decision-making process in the legislative and executive bodies. Primarily seen as a compromise to bring back inter-ethnic trust, in recent years, it has been increasingly seen by European Institutions such as European Court of Human Rights and the European Commission, as discriminatory and ineffective, hampering the progress of the state on its path to EU integration. The European Court of Human Rights has in multiple cases described the realization of principle of constituent peoples as contrary to human rights standards. On the other hand, the EU institutions, such as European Commission, in its assessments of the readiness of Bosnia and Herzegovina to progress in EU integrations and potentially become a EU Member state, has pointed to the fact that realization of the principle is detrimental to the decision-making process. The non-compliance with the European Court of Human Rights decisions related to the discriminatory nature of the principle, has slowed the country’s EU integration process, even in the stage of signing of the Stabilization and Association Agreement, as is still an outstanding obligation.  All of that is negatively impacting institutional frameworks aimed at devising and implementing necessary reforms on the EU integration path. This article analyses the position of the principle of “constituent peoples” in the legal system of the country and its evaluation by the European institutions as detrimental to country’s progress.

Every day, new discoveries are made by researchers from all across the globe and fields. HICSS is a flagship venue to present and discuss such scientific advances. Yet, the activities carried out for any given research can hardly be fully contained in a single document of a few pages-the"paper."Indeed, any given study entails data, artifacts, or other material that is crucial to truly appreciate the contributions claimed in the corresponding paper. External repositories (e.g., GitHub) are a convenient tool to store all such resources so that future work can freely observe and build upon them -- thereby improving transparency and promoting reproducibility of research as a whole. In this work, we scrutinize the extent to which papers recently accepted to HICSS leverage such repositories to provide supplementary material. To this end, we collect all the 5579 papers included in HICSS proceedings from 2017-2024. Then, we identify those entailing either human subject research (850) or technical implementations (737), or both (147). Finally, we review their text, examining how many include a link to an external repository-and, inspect its contents. Overall, out of 2028 papers, only 3\% have a functional and publicly available repository that is usable by downstream research. We release all our tools.

Large Language Models (LLMs) are transforming human decision-making by acting as cognitive collaborators. Yet, this promise comes with a paradox: while LLMs can improve accuracy, they may also erode independent reasoning, promote over-reliance and homogenize decisions. In this paper, we investigate how LLMs shape human judgment in security-critical contexts. Through two exploratory focus groups (unaided and LLM-supported), we assess decision accuracy, behavioral resilience and reliance dynamics. Our findings reveal that while LLMs enhance accuracy and consistency in routine decisions, they can inadvertently reduce cognitive diversity and improve automation bias, which is especially the case among users with lower resilience. In contrast, high-resilience individuals leverage LLMs more effectively, suggesting that cognitive traits mediate AI benefit.

Early elementary education represents a critical period for the development of foundational reading and mathematics skills. This study investigated the relative contributions of teacher ratings, cognitive abilities, fine motor skills, and gender to academic achievement in a sample of 129 primary school students (Grades 2–5; M age = 9.1 years, SD = 1.2; 68 girls). Reading and mathematics performance were assessed using decoding and arithmetic tasks, respectively. Predictor variables included teacher ratings of academic performance, processing speed, visual-spatial working memory, fine motor speed, visuomotor integration, and gender. Hierarchical regression analyses revealed that teacher ratings were the strongest predictors of reading achievement but were less predictive of mathematics outcomes. Processing speed significantly predicted both reading and mathematics performance. Working memory showed no significant effect on mathematics and a negative association with reading after motor skills were included in the model, suggesting a suppression effect. Fine motor skills emerged as unique contributors to reading, but not mathematics, highlighting their specific relevance for literacy acquisition. Gender differences were observed only in mathematics, where boys outperformed girls; no significant gender effect was found for reading. These findings highlight the importance of integrating teacher evaluations with cognitive and motor assessments to gain a better understanding of early academic achievement.

BACKGROUND Selection of acute stroke patients for endovascular thrombectomy (EVT) within 6 h from symptom onset can be done using MRI or CT. However, association of either imaging modality with better clinical outcomes or shorter workflow times is still not fully understood. METHODS We searched Medline and Ovid-Embase for studies comparing outcomes and workflow metrics between patients selected for EVT using CT or MRI from inception to November 30, 2024. The primary outcome was the association of imaging modality with functional independence (modified Rankin Scale score, 0-2) at 90-days and workflow metrics. Pooled odds ratios with 95% CIs were calculated using a random-effects model. RESULTS Nine studies with 11,202 patients (3018 with MRI vs 8184 with CT) were analysed. Patients selected with MRI had similar odds of 90-day mRS 0-2 (adjusted odds ratio [aOR] 1.1195% CI 0.84-1.47) and lower mortality (aOR 0.6695% CI 0.57-0.76) compared to those selected with CT. Door-to-imaging time (mean difference [MD] 11.2 min 95% CI 4.8 to 18.4) and door-to-intravenous thrombolysis initiation time (MD 10.1 min 95% CI 4.9 to 15.2) were longer in patients selected with MRI. However, door-to-arterial puncture time was similar between both groups (MD 6.8 min 95% CI -4.6 to 18.1). CONCLUSION In stroke patients undergoing EVT within 6 h from symptom onset, MRI before EVT could be a feasible alternative to CT without significant delays from door to puncture time. Randomized trials are needed before these findings can be generalized.

The specificity of the business of agro-food companies is that their products have little or no impact on the environment. However, environmental pollution of these products is caused by the use of packaging. Therefore, it is necessary to apply the principles of the circular economy in the business of companies. Applying green packaging that has little or no impact on the environment helps in preserving the environment. Companies usually purchase packaging from suppliers and therefore, it is necessary to choose the right supplier from which to purchase green packaging to support the implementation of the circular economy. The aim of this research is to select a green packaging supplier for company X in order to influence the development of a circular economy in the company’s business. Based on this, the following research question is considered in this paper: how can the selection of a green packaging supplier influence the implementation of a circular economy at company X? The research covers ten criteria used in this selection, with which eight suppliers were observed. Because every decision-making process in the economy is characterized by risk and insecurity that affects the uncertainty in decision-making, an intuitionistic fuzzy set (IFS) was used. Determining the importance of weights was performed directly based on the ratings of the decision-maker (DM) and the steps of the SiWeC (Simple Weight Calculation) method, as well as using the Entropy method. The compromise results of these methods showed that the most important criteria for assessing the life cycle of packaging are transparency and ethics in business. The ranking of suppliers was carried out using the TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) method and its results showed that supplier 5 is the first choice for establishing long-term cooperation in the procurement of green packaging.

Introduction The 2019 ESC/EAS guidelines introduced stricter low-density lipoprotein cholesterol (LDL-C) targets, particularly for patients at high and very high cardiovascular (CV) risk. However, data on the implementation of these targets in real-world clinical practice—especially in countries with high/very high CV risk—remain limited. The DISCOVERY study aimed to assess LDL-C management, lipid-lowering therapy (LLT) use, and guideline adherence across multiple countries in Central and Eastern Europe and Central Asia. Methods This prospective, observational, multicenter study enrolled adult patients with hypercholesterolemia (HCL) from 10 countries grouped into three regions. Data was collected at baseline and after 12 weeks of follow-up. LLT patterns, LDL-C levels, target attainment (both investigator-defined and 2019 ESC/EAS-recommended), and physician adherence to guidelines were analyzed. Results A total of 6,447 patients were included; 53.2% were female, and the mean age was 60.5 ± 11.9 years. Most patients (66%) were in secondary prevention. At baseline, 36.8% had been treated with LLT. After the first visit, treatment was changed in 78% of patients, but only 42.4% received high-intensity statins and 9.3% received statin-ezetimibe combinations at follow-up. LDL-C target achievement was poor: only 5.6% of patients met the guideline-recommended LDL-C goals, compared to 45.5% who met physician-defined targets. Among patients with ASCVD, only 3.3% achieved guideline LDL-C targets. The most significant gap was observed between guideline recommendations and physician-set LDL-C goals. No significant difference in LDL-C target attainment was observed between specialists and general practitioners. Discussion The DISCOVERY study reveals suboptimal LDL-C control and low adherence to the 2019 ESC/EAS guidelines in routine practice across countries with high/very high CV risk. These findings highlight the urgent need for strategies to improve physician awareness, promote intensive LLT use, and close the gap between guidelines and clinical practice. A paradigm shift toward proactive LDL-C management is essential to reduce residual CV risk in these populations.

The historical facts of the development of the information/communication and media system in Bosnia and Herzegovina represent that each political system has reformed the existing communication channels, organizing them according to its own ideological ambitions. Different types and forms of social control of the media space in Bosnia and Herzegovina in history have most often been motivated by “fear of political subversion“ (McQuail, 2010). The contemporary history of the nineties and the war in Bosnia and Herzegovina resulted in the creation of media spaces that corresponded with the ethnically divided territory and continued the media policies of interpellation, blind following/manipulation of the ideology that Marxist philosopher Althusser defines by imposing ways of thinking or living that correspond to established national concepts. Forms of political parallelism of the media in Bosnia and Herzegovina today are visible through indicators of organizational or financial connections between media structures and different political/interest groups, political orientation of editorial and journalistic staff, and the use of media conditioned by territorial-ethnic criteria. The subject of the analysis is the degree and form of political parallelism of the Bosnian-Herzegovinian media in history. The analysis tried to answer the problematic question related to the issue of objective recognition of the needs and interests of the public in the creation of everyday media agendas, which, among other things, should be in the function of criticism, control and integration, and which are today strongly influenced by historical media practice and media heritage. The research starts from the hypothesis that every state-legal organization in the history of Bosnia and Herzegovina, in order to establish and consolidate a certain system, also established specific communication/media structures, which enabled media promotion and consolidation of the already existing socially established practice of an ethnically divided society. The aim of the research is to point out the importance of establishing and developing stable and independent media organizations in Bosnia and Herzegovina that will enable the nation to imagine itself as coherent, significant and homogeneous community and to point out that basic media functions can be valuable resources for overcoming ethnic, political, nationalist and hegemonic policies in contemporary social constellations.

The Institute for Literary and Cultural Studies of the NIRSA Center of the University of Sarajevo – Faculty of Philosophy, and the Department of Comparative Literature and Information Sciences, University of Sarajevo – Faculty of Philosophy, organized on 19 April 2024 an international scientific conference "Engaged Art", held at the University of Sarajevo – Faculty of Philosophy. Seventeen authors from Bosnia and Herzegovina, Slovenia, Croatia, and Montenegro participated in this scientific conference. We are pleased to publish nine scientific papers that were developed after presentations at this gathering.

The aim of this paper is to try to summarize and problematize, and then to offer some theses that were discussed at the Scientific Meeting entitled Engaged Art. The gathering was held on April 18, 2024, in the premises of the Faculty of Philosophy in Sarajevo. Eighteen speakers participated in this Conference. Approaches to this topic were interdisciplinary (fine art, novel, poetry, theatre, film, drama, etc.), historically different (contemporary, ancient Greece, 19th century, etc.), so the following questions were highlighted, which this text underlines: what is engaged art today anyway?, is art in itself already engaged or is there really a special genre of "engagement"?, is engaged art connected to ideology, and if so, in what way!? These and similar questions are the subject of this work.