Publication Search

Object detection plays a crucial role in intelligent transportation systems, particularly for outdoor traffic monitoring applications that require accurate and real-time performance under limited computational resources. Recent developments in YOLO-based architectures have introduced multiple model variants; however, their practical performance under constrained training conditions remains insufficiently explored. This study presents a comparative evaluation of YOLOv5, YOLOv7, and YOLOv8 for outdoor traffic object detection using a real-world dataset and identical experimental settings. The main objective of this research is to analyze the robustness and detection quality of different YOLO variants when trained with a limited number of epochs, reflecting practical deployment scenarios. All models were trained and evaluated using the same dataset, preprocessing pipeline, and hardware configuration to ensure a fair comparison. Performance evaluation was conducted using multiple metrics, including precision, recall, mAP@50, Precision–Recall curves, area under the curve (AUC), and peak F1-score. Experimental results indicate that YOLOv5 outperformed YOLOv7 and YOLOv8 in terms of overall detection stability and robustness. The merged Precision–Recall analysis shows that YOLOv5 achieved a higher effective AUC and superior mAP@50, reflecting better global detection performance. In addition, YOLOv5 exhibited a higher peak F1-score, indicating a more balanced trade-off between precision and recall. In contrast, YOLOv7 and YOLOv8 showed performance degradation under limited training conditions despite their more advanced architectures. These findings suggest that YOLOv5 remains a reliable and efficient solution for outdoor traffic object detection, particularly in resource-constrained environments. The study highlights the importance of comprehensive evaluation metrics and practical experimental settings when selecting object detection models for real-world applications.

Healthcare institutions worldwide are undergoing digital transformation to improve efficiency and patient experiences. While many studies focus on clinical applications of information technology (IT), its impact on administrative service quality remains limited, even though administrative services are patients’ primary point of contact. This study examines how IT adoption influences administrative service quality in Indonesian healthcare institutions, focusing on relationships between IT implementation levels and service quality dimensions, as well as mediating and moderating factors. Using a mixed-methods sequential explanatory design, quantitative data were collected from 385 patients and 127 administrative staff across 24 hospitals, supported by 32 in-depth interviews. Service quality was measured using SERVQUAL dimensions: tangibles, reliability, responsiveness, assurance, and empathy. The findings show significant positive relationships between IT adoption and all service quality dimensions, with the strongest effects on reliability and responsiveness. Staff digital competency and system usability partially mediate these relationships, while implementation quality acts as a key moderating factor. Well-implemented systems yield substantially higher service improvements than poorly implemented ones. Most patients prefer digital services when functioning properly, although many report frustration when systems fail or staff lack competency. This study highlights the importance of effective IT implementation to enhance administrative service quality and offers practical insights for healthcare management.

Education involves acquiring knowledge and skills, with English learning being a skill-based process that requires consistent and engaging practice. In the digital era, mobile applications have emerged as effective tools to overcome traditional limitations in language learning. This study investigates the effectiveness of the Duolingo application in enhancing English vocabulary among university students. A pre-experimental one-group pretest-posttest design was employed involving 40 first-semester Pharmacy students at Anwar Medika University. Participants engaged in daily Duolingo practice (20 XP per day) for 30 consecutive days. Data were collected through vocabulary pretest and posttest as well as a Likert-scale questionnaire. Quantitative analysis revealed a significant improvement in vocabulary scores, with the mean score increasing from 57.25 in the pretest to 81.70 in the posttest. Questionnaire results indicated high student motivation, positive perceptions of the app’s usefulness and ease of use, and minimal perceived disadvantages. The findings suggest that daily Duolingo practice is an effective, engaging, and accessible method for boosting English vocabulary acquisition. This study provides practical insights for educators seeking innovative technology-integrated approaches to language teaching.

Attention mechanisms have been widely incorporated into recurrent neural network architectures for financial time series forecasting, with most prior work reporting improvements in price-level error metrics. This study revisits that claim through a controlled empirical comparison of four deep learning architectures on nearly two decades of Telkom Indonesia (TLKM) closing price data from the Indonesia Stock Exchange (IDX). The models evaluated are a three-layer Gated Recurrent Unit (GRU) baseline, a comparable Long Short-Term Memory (LSTM) network, a Bahdanau end-attention GRU (Attn-GRU-V2), and a multi-head self-attention GRU hybrid (Attn-GRU-V3). Each architecture is trained over 30 independent runs with distinct random seeds, and performance is reported as 95% confidence intervals derived from the t-distribution. Statistical comparisons employ the Wilcoxon signed-rank test, a nonparametric paired test appropriate given the confirmed non-normality of residuals. The main finding is a consistent trade-off: the plain GRU achieves the lowest RMSE (94.02 ± 1.22 IDR) across all 30 runs, while Attn-GRU-V2 achieves the highest directional accuracy (45.91 ± 0.09%), surpassing GRU in every independent run. Bahdanau attention weights are nearly uniform across the 30-day lookback window (coefficient of variation: 3.21%), indicating that the mechanism cannot identify selectively informative timesteps in this univariate price series. This finding is consistent with the weak-form Efficient Market Hypothesis for the Indonesian market. An ablation study reveals that a 20-day lookback window maximizes directional accuracy (47.72 ± 0.21%) for the Attn-GRU-V2 model. These results suggest that Bahdanau end-attention consistently and significantly improves directional accuracy relative to a plain GRU baseline, providing an architecturally attributable advantage for direction-based applications, even when absolute price-level error is not reduced. The directional accuracy values remaining below 50% across all models are consistent with a weak-form efficiency characterization of the Indonesian market.

The increasing sophistication of malware has rendered traditional signature-based detection methods insufficient, necessitating behavior-driven and adaptive analytical frameworks. This study presents a sequential deep learning framework that models system-level API call sequences as structured linguistic representations for behavioral malware detection. Unlike conventional comparative studies, this work systematically evaluates recurrent and attention-based architectures under controlled experimental conditions, with a particular focus on generalization performance and overfitting mitigation. Two neural architectures, a Long Short-Term Memory (LSTM) network and a Transformer-based attention model, are trained on publicly available API call sequence data for binary classification of malicious and benign executables. Beyond standard accuracy metrics, the study further examines model stability, convergence behavior, and the impact of long-range dependency modeling on detection robustness. Experimental results demonstrate that the Transformer architecture achieves superior performance, attaining 95.54% classification accuracy and consistent improvements in precision, recall, and F1-score, indicating a stronger ability to capture complex behavioral dependencies. These findings highlight the effectiveness of attention mechanisms in behavioral malware modeling and provide empirical evidence that NLP-inspired architectures offer a robust and scalable approach for real-world cybersecurity applications.

The principle of justice constitutes the core of debates regarding the permissibility of polygamy in Islamic law; however, the standards used to assess justice in judicial practice still vary between countries. This study comparatively examines how the Malaysian Syariah Court and the Indonesian Religious Court implement the concept of justice as the primary requirement for granting polygamy permits in judicial practice. The research employs a comparative legal approach and normative-juridical analysis of court decisions and statutory regulations applicable in both countries. The findings reveal that although both judicial systems refer to Qur’an Surah An-Nisa verse 3 as the normative foundation, significant differences exist in the mechanisms for proving justice, the consideration of the interests of existing wives, and the role of judges in assessing the feasibility of polygamy applications. Malaysia tends to apply a more structured standard of justice through strict technical regulations, while Indonesia provides broader judicial discretion by considering sociological aspects. Both countries also face similar challenges in translating immaterial justice into objective, consistent, and measurable legal decisions.

This study aims to analyze the implementation of e-government in improving the quality of public services at the Population and Civil Registration Office of Gresik Regency. The study uses a qualitative descriptive approach, with data collection techniques including observation, interviews, and documentation. The analysis focuses on the stages of e-government implementation, covering the dimensions of presence, interaction, and transaction in population administration services. The results indicate that the implementation of e-government at the Population and Civil Registration Office of Gresik Regency has been carried out through the provision of digital information media, such as an official website and online-based population administration service applications. In the presence dimension, the local government provides various information related to population administration services, including requirements, procedures, service times, and the types of services available to the public. In the interaction dimension, the digital service system allows the public to communicate with the service office by submitting questions, complaints, or requests for information online. Meanwhile, in the transaction dimension, the public can submit requests for population documents, such as Family Cards, birth certificates, and other documents, through the digital service system. The implementation of e-government has positively impacted the efficiency, transparency, and ease of access to population administration services for the public. Therefore, the utilization of information technology in public services can serve as an important strategy for improving the quality of population administration services in local government.

Digital transformation in the public sector has significantly driven service innovations, particularly in pension administration for Civil Servants (ASN). This study aims to analyze the development of digital-based public service innovations in pension administration while identifying the benefits and challenges associated with their implementation. The research employs a Systematic Literature Review (SLR) method by examining relevant scholarly articles from national journals focusing on the digitalization of public services and pension systems. The literature selection process was conducted systematically to identify, evaluate, and synthesize key findings related to digital pension service innovations. The results indicate that digitalization through applications and electronic platforms enhances administrative efficiency, accelerates data verification, and improves the speed of pension fund disbursement. It also strengthens transparency and accountability while simplifying bureaucratic procedures and expanding service accessibility for retirees. However, several challenges remain, including low digital literacy among retirees, limited access to technological devices, and insufficient public awareness regarding digital service usage. The findings suggest that the success of digital-based public service innovations depends not only on technological availability but also on human resource readiness, institutional capacity, and the level of public acceptance. Therefore, a comprehensive strategy is required, including improving digital literacy, strengthening information technology infrastructure, and optimizing communication efforts to ensure effective adoption.

Digital transformation in the public sector has driven various service innovations, including pension administration services for State Civil Apparatus (ASN). This study aims to analyze the development of digital-based public service innovations in pension administration and identify the benefits and challenges of their implementation. This study used the Systematic Literature Review (SLR) method by reviewing various relevant scientific articles from national journals that discuss the digitalization of public services and pension administration. The literature selection process was carried out systematically to identify, evaluate, and synthesize research findings related to digital-based pension service innovations. The review results indicate that digitalization of pension administration services through the use of electronic service applications and platforms can improve administrative efficiency, accelerate data verification and pension fund disbursement, and increase transparency and accountability in public services. The implementation of digital services also contributes to simplifying bureaucratic procedures and increasing service accessibility for retirees. However, the literature review also revealed challenges in implementing digital pension services, including low digital literacy among retirees, limited access to technological devices, and suboptimal dissemination of service information. The findings of this study indicate that the success of digital-based service innovations depends not only on technology, but also on human resource readiness, the organizational capacity of government institutions, and the level of public acceptance of the use of digital technology. Therefore, developing digital-based pension services requires a comprehensive strategy.

This research aims to analyze the urgency of implementing Statement of Use, Declaration of Use, and Specimen of Use instruments as a means of supervising trademark use obligations in Indonesia. Currently, Law Number 20 of 2016 concerning Trademarks and Geographical Indications (UU MIG) only requires a written statement of use without being accompanied by concrete evidence at the time of renewal applications. This creates a loophole for trademark registrations without the intent to use (non-use), which can hinder bona fide applicants.  This study employs a normative legal research method with a comparative law approach between the UU MIG and the Lanham Act (U.S. Trademark Act of 1946). The Lanham Act mandates proof of actual trademark use through a Statement of Use for intent-to-use applications, as well as a Declaration of Use for applications based on use-in-commerce, which must be submitted with a Specimen of Use as actual evidence of the trademark's use in trade.  The results indicate that the adoption of these instruments into the Indonesian legal system is urgent to strengthen the supervisory function of the Directorate General of Intellectual Property (DGIP). This implementation would enable the DGIP to conduct administrative (ex-officio) trademark cancellations for non-use, thereby purging the trademark database of passive marks without having to rely on third-party lawsuits. This legal reform is in line with the legal development theory to encourage business actors to actively utilize the economic functions of trademarks.

This qualitative literature review explores the role of behavioral nudging as a tool for enhancing corporate ethics and curbing organizational misconduct. Drawing on empirical and theoretical studies from behavioral ethics, organizational psychology, and compliance management, the review examines how subtle interventions—such as reminders, social norm cues, and visual prompts—can influence ethical decision-making in the workplace. The findings suggest that nudges are effective in reducing ethical fading, improving compliance, and reinforcing ethical culture when aligned with organizational values and context. However, concerns regarding manipulation, cultural adaptability, and long-term efficacy remain. This review highlights the importance of integrating nudging within a broader ethical infrastructure and calls for future research on scalable, transparent, and culturally sensitive applications of ethical influence in diverse organizational settings

This community service activity aims to enhance students’ competencies in business management, production, and operations through an industrial visit to PT. XYZ. The activity involved 65 students from the Entrepreneurship Study Program of Universitas Negeri Medan and was conducted in December 2025. The method used was experiential learning, allowing students to gain direct exposure to real industrial practices. The stages of the activity included preparation, implementation of the industrial visit, observation and data analysis, and evaluation and reflection. The results of the activity indicate significant improvements in five key areas: understanding of business management, production processes, operations management, development of soft skills, and increased entrepreneurial motivation and interest. Students were able to connect theoretical knowledge with real-world applications, thereby enhancing their analytical and practical abilities. In addition, the activity contributed to the development of communication skills, teamwork, critical thinking, and professional attitudes. Overall, this program demonstrates that industrial visits are an effective learning strategy in entrepreneurship education to produce competent, adaptive, and competitive graduates.

This community service activity aims to enhance the capacity of Micro, Small, and Medium Enterprises (MSMEs) in preparing financial statements based on the Financial Accounting Standards for Micro, Small, and Medium Entities (SAK EMKM) through the digitalization of financial recording in UMKM Pasadana. The main problems faced by the partners include low accounting literacy, unstructured and simple financial recording practices, and the limited utilization of digital technology in financial management. The implementation method adopts a participatory approach consisting of several stages, namely socialization, training, mentoring, and evaluation. The training focuses on understanding the basic concepts of SAK EMKM and the practical use of simple and applicable digital financial recording applications. Mentoring is conducted directly to ensure that MSME actors are able to implement financial recording independently and sustainably. The results of the activity indicate an improvement in the knowledge and skills of MSME actors in conducting systematic financial recording in accordance with SAK EMKM principles. In addition, MSME actors begin to adopt digital technology in recording transactions, which improves efficiency, accuracy, and transparency of financial statements. This activity also encourages a shift in mindset, where financial recording is no longer viewed as merely administrative, but as a basis for business decision-making. Thus, this community service activity provides a practical contribution to improving the quality of MSME financial management through the integration of accounting literacy and sustainable digital technology utilization.

The digital era, with 78% internet penetration in Indonesia (2025), brings information advancement but also threats like cyberbullying, hoaxes, and SARA polarization through social media. This study aims to analyze the relevance of Pancasila values as an ethical filter in mitigating these negative digital impacts through social media usage case studies. Employing a qualitative approach based on library research, data was gathered from 18 Sinta-accredited journals (2021-2026), 2 Pancasila digital theory books, UU ITE regulations, and APJII reports. Content analysis with Miles & Huberman (2024) data reduction was applied to code the implementation of each Pancasila principle. Results show that the first principle combats religious intolerance, the second suppresses cyberbullying (25% reduction), the third reduces 2024 election polarization (40%), the fourth promotes digital deliberation, and the fifth closes rural literacy gaps through gotong royong crowdfunding (Rp1T collected). Viral disinformation and Lombok 2025 disaster cases prove Pancasila's effectiveness beyond formal regulations. It is concluded that Pancasila is adaptive as a moral algorithm in the digital era, transforming social media from conflict breeding grounds into national integration spaces. Recommendations include strengthening the "Pancasila Digital Ethics" curriculum for Gen Z/Alpha, national AI literacy applications, and platform collaboration with BPIP-Kominfo.

This study aims to analyze the Transformation of E-Government Towards E-Governance in the Public Service Process at the Population and Civil Registration Office of Gresik Regency. The utilization of information technology in public services is one of the government’s efforts to improve administrative efficiency, transparency, and the quality of services provided to the public.This research employs a qualitative approach with a descriptive method to provide an in-depth depiction of the implementation of E-Governance in population administration services. Data collection techniques include interviews, observations, and documentation, while data analysis follows the stages of Grouping the data according to key constructs, Identifying bases for interpretation, Developing generalizations from the data, Testing alternative interpretations, and Forming and/or refining generalizable theory from the case study.The results indicate that the implementation of the Transformation of E-Government Towards E-Governance in the Public Service Process at the Population and Civil Registration Office of Gresik Regency is carried out through several key dimensions, namely E-Administration, E-Service, and E-Society. The E-Administration dimension is reflected in the use of digital-based administrative systems for managing population data and processing document applications electronically. The E-Service dimension shows that online services provide easier access for the public to manage population documents more quickly and efficiently. Meanwhile, the E-Society dimension demonstrates an increased utilization of information technology by the public in accessing population administration services. Nevertheless, the implementation of digital services still faces several challenges, such as limited digital literacy among the public and uneven internet access. This study concludes that the application of E-Governance in population administration services in Gresik Regency has made a positive contribution to improving the quality of public services through the utilization of information technology.

This systematic literature review analyzes previous studies on impact of educational technology on students’ behavioural engagement in English language learning. The integration of educational technology in English classroom has increased significantly in order to address students’ low participation, limited attention, and passive learning behaviour. Various digital tools such as digital storybooks, digital storytelling, multimedia learning platforms, and interactive online applications have been implemented to promote students’ active participation. This review followed the prefered Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework to ensure a systematic and transparent review process. Relevant journal articles published between 2020 until 2025 were identified through Google Scholar, ResearchGate, and Publish or Perish. A total of 15 articles met the conclusion criteria and were analysed in this study. The findings indicate that educational technology positively influences students’ behavioural engagement, which is reflected in increased participation, attention, interaction, and task completion in English leraning contexts.The review higlights that the appropriate intergration of educational technology can support more interactive and learner-centred English language learning environments.

Implementing REST in modern applications, security will be a key foundation for its development because the REST architecture requires communication between servers. In this study, we will enhance REST request security by using SHA-1 tokens and the Keccak algorithm. Tokens are the access keys for making requests. The token generation process is carried out on the server; the client will generate a token, and the server will return a valid token. This valid token can be used to request data from the server. Adding a token will impact the security and speed of REST. The token will be verified by the server and declared valid. If valid, the server will return the data; otherwise, the server will send an error message. Compared to using a token, data security is more assured. Furthermore, adding a token parameter will increase the token verification process, thus increasing the number of processes, which will impact speed. The results of this test show that server data security is better maintained and more secure compared to using a token, because anonymous users cannot access the data. The API access speed without a token is 48.8 milliseconds, while using a SHA-1 token is 62.3 milliseconds, and the Keccak algorithm is 58.9 milliseconds. The time efficiency reduction for implementing the SHA-1 token algorithm is 27.67% or 13.5 milliseconds, and the Keccak algorithm is 26.6% or 10 milliseconds.

This study investigates the role of artificial intelligence (AI) in enhancing language learning, with a focus on five key applications: automatic text analysis, personalized learning, adaptive feedback, language error detection, and automatic translation. The study addresses the challenge of integrating AI effectively in educational contexts while balancing technological potential with pedagogical guidance. The objective is to provide a comprehensive understanding of how AI tools contribute to more adaptive, efficient, and engaging language learning experiences. A systematic literature review method was employed, selecting and critically analyzing studies published between 2020 and 2025 that examined AI-assisted language learning strategies. The findings indicate that automatic text analysis supports comprehension monitoring and guided learning, while personalized learning adapts content to individual learner needs, enhancing motivation and retention. Adaptive feedback delivers immediate, targeted guidance that fosters accuracy and self-regulated learning, and language error detection tools enable learners to identify and correct grammatical and lexical mistakes, promoting metalinguistic awareness. Automatic translation broadens access to authentic texts and cross-cultural materials, supporting comprehension and independent learning. Synthesizing these findings highlights the transformative potential of AI to improve learning outcomes while also revealing challenges such as tool reliability, ethical considerations, and the need for teacher oversight. The study concludes that AI, when thoughtfully integrated, complements instruction, enhances learner engagement, and supports differentiated and data-driven teaching strategies, providing valuable insights for language educators and guiding future research on AI-enabled language learning.

This study aims to evaluate the performance of digital information systems, including websites and mobile applications, used by national expedition companies in Indonesia, using standardized technical indicators from Google Web.dev. This research employs a descriptive and evaluative approach with purposive sampling. The research objects comprise five major national expedition companies: JNE, J&T Express, SiCepat, Pos Indonesia, and Lion Parcel. The evaluation is conducted using four main indicators: Performance, Accessibility, Best Practices, and Search Engine Optimization (SEO). The results indicate that, in general, the digital information systems of national expedition companies function adequately; however, there are notable performance variations among companies, particularly in terms of loading speed and mobile performance. Pos Indonesia demonstrates the most balanced overall performance, while several other companies require further optimization, especially in performance efficiency and SEO. This study contributes to the literature by providing an objective technical evaluation of expedition information systems and offering practical recommendations to improve digital service quality, enabling faster, more secure, and more responsive logistics services in a highly competitive industry.

Explainable artificial intelligence (XAI) has become a critical requirement in cybersecurity due to the high-stakes nature of security decision-making and the limitations of black-box learning models. This study investigates the construction of an explainable cybersecurity knowledge representation by leveraging standardized terminology from the NIST cybersecurity glossary. The primary problem addressed is the lack of transparent and semantically grounded reasoning mechanisms in existing AI-driven cybersecurity systems, which limits trust, accountability, and analyst adoption. To address this challenge, we propose a NIST-based semantic knowledge graph that embeds explainability directly into its ontology structure and reasoning process. The proposed framework systematically extracts definitional entities and relations from NIST glossary entries to construct a domain ontology and a multi-relational knowledge graph. A rule-based semantic relation extraction method is employed to ensure faithful, interpretable, and reproducible reasoning paths. The resulting knowledge graph contains over 3,000 cybersecurity concepts and approximately 27,000 semantic relations, covering hierarchical, associative, dependency, and mitigation semantics. Experimental evaluation demonstrates that the proposed approach achieves a high level of explainability, with 92.4% of reasoning outcomes being fully traceable and only 1.4% classified as non-traceable. Most explainable reasoning paths are limited to two or three hops, indicating an effective balance between inferential depth and human interpretability. Structural analysis further confirms the presence of meaningful hub concepts that support multi-hop semantic inference. These results confirm that ontology-driven, standard-based knowledge graphs provide a robust foundation for explainable cybersecurity intelligence. The study concludes that explainability-by-design, grounded in authoritative standards, offers a viable and trustworthy alternative to opaque AI models for cybersecurity applications.