Publication Search

This study conducts sentiment analysis of Indonesian user reviews of the CapCut application using IndoBERT and compares two evaluation schemes: a single 80/20 train–test split and stratified 5-fold cross-validation (k=5). A total of 1,048,575 reviews were collected from the Google Play Store through web scraping and labeled into three sentiment classes based on rating: negative (1–2), neutral (3), and positive (4–5). After preprocessing—cleaning, case folding, banned-word removal, normalization—and duplicate removal, 517,962 reviews were retained. IndoBERT Base P1 was fine-tuned using fixed hyperparameters (batch size 32, learning rate 2e-5, up to 4 epochs, early stopping patience 2), while undersampling was applied to the training set to address class imbalance. Performance was assessed using accuracy, precision, recall, F1-score, and ROC-AUC, supported by confusion matrix and ROC-curve visualizations. The single split achieved an accuracy of 0.756, whereas cross-validation produced a mean accuracy of 0.740. Across both schemes, the positive class achieved the best performance (F1-score 0.850; ROC-AUC 0.918–0.919), while the neutral class remained the most challenging (precision 0.198–0.206; F1-score 0.280–0.283). Overall, cross-validation is recommended for reporting because it reduces dependence on a single partition and provides a more representative estimate across multiple splits.

The rapid growth of e-commerce platforms in Indonesia, particularly Tokopedia, has resulted in a large volume of consumer reviews containing valuable information regarding customer perceptions and satisfaction. However, manual analysis of such reviews is inefficient and prone to subjectivity, necessitating an automated approach based on machine learning. This study aims to classify the sentiment of sports product reviews on Tokopedia into positive, negative, and neutral categories by applying Logistic Regression, Support Vector Machine (SVM), and Random Forest using the Term Frequency–Inverse Document Frequency (TF-IDF) approach. The data were collected through web scraping of Indonesian-language sports product reviews and processed through several preprocessing stages, including data cleaning, case folding, tokenization, stopword removal, and stemming. Feature representation was performed using TF-IDF to transform textual data into numerical vectors, after which the dataset was divided into training and testing sets with an 80:20 ratio. Model performance was evaluated using accuracy, precision, recall, and F1-score metrics. The results indicate that the application of TF-IDF significantly improves the performance of all models, with SVM consistently achieving the most optimal performance compared to Logistic Regression and Random Forest. These findings demonstrate that classical machine learning algorithms combined with TF-IDF remain highly effective for sentiment analysis of Indonesian-language text. The implications of this study are expected to assist sellers in understanding customer opinions, support consumers in making informed purchasing decisions, and serve as a foundation for the development of sentiment analysis and recommendation systems on e-commerce platforms.

This study examines Japanese advertising and product-information texts on Shiseido Japan’s official website (www.brand.shiseido.co.jp) that grammatically prevent readers from construing statements as universal claims (“always” or “true for everyone”). It addresses two problems: how universal readings are blocked through grammatical construction in this register, and how the main blocking mechanisms differ in limiting generalisation and managing scope. The data consist of sentence-level usage, precautionary, and quality-related statements that plausibly invite broad general interpretations. Seven analytically representative tokens are used as illustrative evidence, covering wake-negation, baai-based case framing, and event/occasion packaging with V-ru koto ga aru, including rare-event calibration with mare ni and layered conditional framing. The study employs qualitative, theory-driven grammatical analysis focusing on clause structure, embedding, nominalisation, connective relations, and the compositional contribution of key markers. The results identify recurring templates with distinct structural signatures. Wake-negation blocks over-strong uptake by denying a candidate inference (…to iu wake de wa arimasen). Case framing with baai shifts categorical commitments into situation-restricted possibility (…baai ga arimasu), including complex variants that add causal linkage, avoidance marking, and directive closure. Event/occasion packaging with koto plus existential predication (…koto ga arimasu) presents anomalies as contingent occurrences, and it can be triggered by causal conditions (e.g., temperature change) or conditional frames (…to). Rare-event marking with mare ni further calibrates frequency and often co-occurs with contrastive reassurance about quality. Overall, universal-blocking emerges as a set of non-redundant grammatical routes that constrain inference, situational domain, and event profiling in a compact public informational genre.

Posyandu (Integrated Health Post) is a public health facility that plays a vital role in providing health services for toddlers and pregnant women. However, data management and reporting often face challenges, such as limited access to information and errors in data recording. Therefore, this study aims to develop a Web-Based Posyandu Payolaumbu Service Information System using the CodeIgniter Framework to improve efficiency and accuracy in data management and reporting. In the development phase, a system requirements analysis and web-based application architecture design were conducted. The system implementation uses the CodeIgniter Framework as a framework to produce a faster, more efficient, and more reliable application. Proposed features include recording health data for toddlers and pregnant women, immunization schedules, weighing, and health reports. The results show that the Web-Based Posyandu Payolaumbu Service Information System can improve efficiency in recording and reporting health data. Users, including posyandu officers, midwives, and administrators, can easily access and manipulate data in real-time. Furthermore, this system helps improve service quality by providing more accurate and complete information on toddler health. In conclusion, the implementation of the Web-Based Posyandu Payolaumbu Service Information System using the CodeIgniter Framework provides significant benefits for data management and health services at Posyandu Payolaumbu. Suggestions for further development include maximizing system utilization, developing additional features, routine maintenance, and ongoing evaluation based on user feedback. With these steps, it is hoped that this system can contribute more effectively to improving the quality of health services at Posyandu and supporting comprehensive public health efforts.

Each country has different holiday policies, but the number of holidays in Indonesia is quite large, which impacts uncertainty for investors when buying or selling shares. These events can cause market anomalies or irregular market conditions and produce abnormal returns at certain times, known as the holiday effect. This study uses a quantitative descriptive method with an event study approach, data collection is carried out using documentation and literature methods. The data used are secondary data in the form of the Jakarta Composite Index (JCI), the LQ45 Index, and the Jakarta Islamic Index (JII) from the official website of the Indonesia Stock Exchange (IDX). Exchange rate data is taken from the official website of Bank Indonesia. The population of this study is every company listed on the IDX, while the data used are JCI, LQ45, and JII data 6 days before and 6 days after the Eid al-Fitr holiday and regular trading days from 2011-2025. The results of the study show that there is no significant difference in the JCI, LQ45 Index, or JII before and after the Eid al-Fitr holiday, so there is no holiday effect. These results indicate that all three indices reflect a market that tends to be efficient and stable in responding to seasonal events. Furthermore, the Rupiah exchange rate had a negative but significant effect on the Jakarta Composite Index (JCI). The Rupiah exchange rate had a negative but insignificant effect on the JII before and after the Eid al-Fitr holiday. The Rupiah exchange rate had a positive but insignificant effect on the LQ45 Index before and after the Eid al-Fitr holiday.

This study is motivated by the high frequency and significant impacts of flooding in Indonesia, particularly the recurrent flooding of the Kali Lamong River in Gresik Regency, which causes substantial socio-economic losses each year. Although disaster management is normatively regulated by national policies, its effectiveness depends largely on collaborative governance among multiple actors. This study aims to analyze the practice of collaborative governance in flood mitigation in Gresik Regency using the model of Weber, Lovrich, and Gaffney (2005), which includes vertical, horizontal, and partnership dimensions, and to identify the key challenges that hinder its implementation. A descriptive qualitative approach was employed in flood-affected areas along the Kali Lamong River, involving BPBD, DPUTR, sub-district governments, village authorities, and local communities through interviews, observations, and document analysis. The findings reveal that collaboration has been established and functions relatively well, particularly in hierarchical coordination, division of roles among government agencies, strengthening of Destana, KENCANA, and SPAB programs, as well as CSR involvement and community participation all contributing to reduced flood duration. However, the implementation of collaborative governance remains constrained by slow land acquisition and infrastructure development, a predominantly reactive orientation, suboptimal early-warning systems between upstream and downstream areas, and uneven support from the private sector and academia. The study concludes that strengthening collaborative mechanisms and accelerating the resolution of structural barriers are essential to achieve more sustainable flood mitigation efforts in Gresik.

This study explores ethical investment practices among Catholic institutions in Indonesia. As a relatively rare area of inquiry, the research investigates the extent to which Catholic institutions have sought to translate the principles of Catholic Social Teaching (CST) into the field of investment. Data were collected through interviews with key participants from three dioceses and six congregations. The findings reveal that these institutions have endeavored to implement ethical principles derived from CST. Due to limitations in their capacity to conduct screening and analysis of various investment options, they have adopted a strategic approach by utilizing the services of an investment consultancy team from the Indonesian Bishops’ Conference (KWI). Drawing on Max Weber’s theoretical framework, this practice is interpreted as a form of social action grounded in “value-rationality.” Thus, investment activities pursued rationally for financial gain are framed within the values of CST. In this way, Catholic institutions have sought to establish faith-based ethical investment practices.

The rapid development of information technology has encouraged organizations to implement computerized systems to improve document management efficiency and security. At PT Petrokimia Gresik, import document archiving was previously conducted using Google Drive and physical storage, which caused issues such as limited access control, risk of data loss, and slow document retrieval during audits. This study aims to design and develop a web-based Import Document Archiving Information System using the waterfall development method. The system was built using HTML, CSS, JavaScript, PHP, and MySQL. The results indicate that the system improves archiving efficiency, enhances data security, accelerates document search, and supports structured monitoring and validation processes.  

The manual archiving process within the Gresik Education Office continues to encounter significant challenges, including difficulties in correspondence tracking, document misplacement, and inefficiencies in disposition management. To address these issues, this study proposes the design and implementation of a Web-Based Archiving and Correspondence Monitoring System developed using the Waterfall methodology. The system aims to facilitate the digital transformation of administrative correspondence through three primary pillars: secure user authentication, flexible digital workflows, and compliance with electronic archival standards. Core features include centralized document management, automated disposition processes, electronic signature verification, and real-time notification integration via WhatsApp. Implementation results demonstrate that the proposed system enhances efficiency, traceability, and accountability in correspondence management at the Gresik Education Office.

The administration of population services in Rumbai Timur District continues to face challenges following the regional expansion that required residents to update their civil documents. The SIPENDUDUK website was introduced to simplify digital submissions; however, its implementation has not been fully effective. A portion of the community, particularly older adults and individuals with low digital literacy, still struggles to navigate service flows and upload required documents. This study aims to evaluate the service quality of the SIPENDUDUK website in improving the effectiveness of civil administration services. A descriptive qualitative method was employed through observations and interviews with officers and service users. The findings indicate that although the website has facilitated administrative processes, its accessibility has not yet reached all user groups due to varying technological capabilities. The system shows adequate efficiency and stability, although several technical issues remain. Service security is maintained through controlled access and a simplified interface. User support is provided through direct assistance from the local office, including manual guidance and mobile outreach services. These results highlight that the success of digital services depends not only on system quality but also on user readiness. This evaluation is expected to contribute to improving the effectiveness, inclusiveness, and overall quality of population administration services in Rumbai Timur District.

Pengujian perangkat lunak merupakan tahap penting dalam memastikan kualitas dan keandalan sistem sebelum digunakan secara luas. Salah satu metode yang umum digunakan adalah black box testing, yaitu pengujian yang berfokus pada fungsionalitas perangkat lunak tanpa memperhatikan struktur internal kode. Dalam penelitian ini, metode black box testing diterapkan dengan menggunakan teknik state transition untuk menguji sistem appointment klinik berbasis web “Citra Medica”. Pengujian dilakukan dengan memetakan alur status pengguna melalui diagram transisi, kemudian membuat skenario uji untuk setiap perubahan keadaan. Hasil pengujian menunjukkan bahwa seluruh fungsi sistem berjalan sesuai dengan spesifikasi, dengan tingkat keberhasilan pengujian mencapai 100% pada setiap skenario transisi. Metode ini terbukti efektif untuk mendeteksi kesalahan logika dan memastikan kestabilan sistem selama perpindahan status pengguna.

Prelabor rupture of membranes (PROM) remains one of the most frequently encountered obstetric complications and continues to contribute substantially to maternal and neonatal morbidity and mortality. Numerous factors have been associated with PROM; however, the available evidence presents considerable variation, highlighting the need to re-evaluate these determinants using recent empirical findings. This review aims to identify and synthesize the principal factors influencing PROM based on ten studies published within the last five years. A systematic search was conducted across PubMed, Scopus, Web of Science, and Google Scholar using predefined keywords, followed by screening through established eligibility criteria. Relevant data from each study including study design, geographical setting, assessed risk factors, and statistical outcomes such as odds ratios or relative risks  were extracted and compared. The review indicates that reproductive and urinary tract infections, inadequate nutritional status, high-risk obstetric history, and short interpregnancy intervals consistently emerge as major determinants. Social factors and the quality of antenatal services were also shown to heighten the likelihood of PROM. Overall, the findings emphasize that PROM arises from a combination of medical and non-medical influences, reinforcing the need for preventive strategies that adopt a comprehensive and integrated approach.

This study discusses the application of the HSV color segmentation method for color-based object detection in digital images. The data used consist of digital images in JPG, PNG, or WebP format containing various colored objects, including red tomatoes, yellow bananas, green apples, orange oranges, purple akebia, brown sapodilla, and blue blueberries. The research process involves converting images from BGR to HSV, determining HSV ranges for each color, creating masks, performing segmentation, analyzing pixels, detecting contours, and visualizing results using bounding boxes. The results show that the HSV method effectively detects objects, separates them from the background, and provides quantitative information, including pixel count, area percentage, and average HSV values for each color. Red, yellow, green, orange, purple, brown, and blue colors were successfully segmented, displaying clear and accurate objects, both for single and multiple objects, under various sizes and lighting conditions. These findings confirm that the HSV method is a simple, fast, and effective approach for color-based image analysis.

Games that were initially enjoyed individually and offline have now evolved into online games that allow interaction, competition, and friendship with other players in real-time. Online games are now increasingly popular, one of which is Mobile Legends. This game can be downloaded through the Play Store and App Store. In April 2024, Mobile Legends ranked the most popular game by downloads on the statista.com website. The research objective was to determine the influence of event marketing, content marketing, influencers, and the social environment on the decision to purchase Mobile Legends game skins in Kebumen Regency. The method used was non-probability sampling with a purposive sampling technique aimed at 100 users. Data collection techniques used questionnaires and literature studies. The data analysis technique used multiple linear regression, which was then processed with the SPSS application version 25.0 for Windows. The results showed that event marketing had a significant effect on purchasing decisions, content marketing had a significant effect on purchasing decisions, influencers had no significant effect on purchasing decisions, the social environment had no significant effect on purchasing decisions, and event marketing, content marketing, influencers, and the social environment simultaneously had a significant effect on purchasing decisions.

Fortuna Furniture still manages its inventory manually, which causes many problems, including delayed reports, inaccurate information, and the potential for lost records. These issues affect the quality of decision-making and operational efficiency. This service aims to build a web-based inventory information system that improves speed, accuracy, and integration in inventory management. The system development process follows the Waterfall model SDLC approach, with requirements analysis, design, implementation, and testing supported by interviews, observation, and documentation. The results of the activity show that the developed system can reduce input errors, speed up the process of recording incoming and outgoing goods, and enable real-time stock monitoring. Additionally, the report generation process becomes more efficient and easier for management to access. Overall, the use of this information system significantly improved the efficiency of Fortuna Furniture's inventory management and demonstrated the importance of digitalization in helping businesses run more smoothly.

The community service program in Purworejo Village, Sragi District, Pekalongan Regency, was implemented to enhance the capacity of local MSME actors in utilizing digital technology as a modern marketing tool. The program adopted a participatory approach involving youth organizations, business actors, village authorities, and the implementation team in every stage of the activity. Assistance included field observation, data collection, content development training, and the creation of a dedicated MSME website as a shared promotional platform. The results indicate significant social changes within the community, including improved digital literacy, stronger awareness of the importance of technology-based marketing, the emergence of local youth as digital facilitators, and the development of collaborative working patterns in managing digital media. These findings reinforce that participatory-based mentoring can strengthen community capacity in adopting technology, consistent with theoretical perspectives that position the community as key agents of local development. Therefore, the program not only produced a functional digital product in the form of an MSME website but also laid the foundation for a more adaptive, competitive, and sustainable village economic ecosystem..

Micro, Small, and Medium Enterprises (MSMEs) play a vital role in the economy of Ketapang Regency but still face challenges in financial recording and management. Many MSME actors have not yet utilized digital technology optimally, leading to manual bookkeeping processes that are prone to errors. This study aims to develop a web-based financial bookkeeping application using the User-Centered Design (UCD) approach, focusing on user needs. The UCD method was applied through four stages: understanding the context of use, specifying user requirements, designing solutions, and evaluating the results. The developed application includes key features such as product management, supplier management, sales recording, receipt printing, and financial reporting. Based on usability testing involving 25 respondents, the application achieved an average satisfaction level of over 85% across aspects of learnability, efficiency, memorability, error handling, and satisfaction. The findings indicate that the application effectively supports MSME actors in recording financial transactions more efficiently, accurately, and reliably. Future improvements may include the integration of digital payment systems, enhanced data security, and interactive graphical financial analysis features.

Although Experience Quality has been extensively researched, comprehensive studies remain limited. This research aims to conduct a Systematic Literature Review (SLR) and bibliometric analysis of existing empirical work on Experience Quality. The methodology used is SLR, with data collected from the Scopus database using the search term "Experience Quality" in the title, abstract, and keywords. This search yielded 1,297 publications from 1965 to 2025 (evaluated on October 24, 2025). The data were analyzed bibliometrically using VOSviewer. The findings confirm that experience quality has a significant effect on brand perception and customer satisfaction, which consequently impacts customer loyalty. Practically, this implies that brand and service managers must treat experience quality as a strategic variable, as good products or services alone are insufficient for long-term loyalty. Socially, positive experiences are shown to strengthen social networks, increase community involvement, and promote inclusion. This research is limited by its reliance solely on the Scopus database; future research should integrate additional databases, such as Web of Science, for broader generalization. Despite the global development of this topic, in-depth, structured studies are scarce. This research is valuable and original as it provides a comprehensive summary and a research roadmap through SLR and bibliometric analysis.

The development of Islamic banking in Indonesia faces challenges in attracting customers, especially younger generations who are more familiar with digital technology. Digital marketing has become an important strategy for reaching a wider market segment through social media, websites, and mobile applications. This study uses a literature review method to analyze the effectiveness of digital marketing in increasing customer interest in Islamic banking. The results of the study show that educational, interactive, and transparent digital marketing can increase customer awareness, trust, and emotional attachment to Islamic products. The factors for the success of this strategy include content quality, ease of access to digital services, content segmentation according to customer profiles, and multichannel communication integration. However, the effectiveness of digital marketing is still limited by the scarcity of educational content, human resource capacity, and the level of digital literacy among the public. This study concludes that optimizing creative, personalized, and educational digital marketing strategies is key for Islamic banking to increase customer interest and loyalty.

This study aims to identify the influence of profitability, liquidity, and asset structure on the capital structure of pharmaceutical sub-sector companies listed on the Indonesia Stock Exchange during the 2019–2023 period. The study spanned five years, from 2019 to 2023. Of the total 15 companies in the population, 7 companies were selected as samples using a purposive sampling method. The research data were sourced from annual financial reports accessed through the official IDX website. Data processing was carried out using multiple linear regression methods. Capital structure was measured using two indicators: the Debt to Equity Ratio (DER) and the Debt to Asset Ratio (DAR). The analysis results showed that profitability had no effect on these two capital structure indicators. Conversely, liquidity and asset structure were shown to influence both DER and DAR. This study provides insight into the factors influencing debt financing decisions in pharmaceutical companies and their implications for the company's financial stability.