Publication Search

Document-level sentiment analysis assigns a single polarity label to an entire review, often obscuring opinion diversity within multi-sentence submissions. This limitation is particularly evident in reviews of multi-service platforms, where users frequently express heterogeneous opinions toward different aspects of the platform in the same review. To address this challenge, this study proposes a sentence-level sentiment analysis framework for Indonesian Gojek app reviews collected from the Google Play Store. The proposed framework introduces a two-stage segmentation strategy that combines punctuation-aware rules with conjunction-aware splitting based on coordinating and adversative conjunctions (e.g., tapi [but], padahal [even though]) to identify opinion boundaries and decompose mixed-sentiment reviews into independently classifiable sentence units. A total of 14,730 raw reviews collected between May and July 2025 were subjected to data cleaning and quality filtering, resulting in 7,187 valid reviews that were further segmented into 14,187 sentence-level instances. Each instance was manually annotated by three annotators using a four-class labeling scheme consisting of app-positive, app-negative, app-neutral, and service categories. Sentiment-level inter-annotator agreement, computed on the subset of instances unanimously categorized as app-related by all three annotators (n = 4,384), achieved substantial agreement (Fleiss'  = 0.636). Hyperparameter optimization was conducted using Optuna with the Tree-structured Parzen Estimator (TPE) sampler across four experimental scenarios. The best performance was achieved by IndoBERTweet under Stratified K-Fold evaluation, attaining an accuracy of 0.751 and a macro F1-score of 0.729, outperforming all IndoBERT configurations. The results demonstrate the effectiveness of domain-adaptive pre-training on informal Indonesian text and highlight the value of conjunction-aware segmentation for preserving fine-grained opinion structures in mixed-sentiment reviews. These findings suggest that domain-aligned language representations provide a practical and effective solution for sentence-level sentiment analysis of Indonesian app reviews.

Breast cancer remains a major global health challenge, with treatment effectiveness often limited by drug resistance and the toxic side effects of chemotherapy on normal cells. The exploration of bioactive compounds from natural sources through a medicinal chemistry approach offers a promising alternative strategy. This study aims to examine the molecular mechanisms of action and Structure-Activity Relationships (SAR) of various natural compound scaffolds as potential breast anticancer agents. The method employed was a systematic narrative literature review of 15 recent scientific articles evaluating computational parameters, including molecular docking, as well as in vitro and in vivo activities. The results indicate that polyphenols, flavonoids such as quercetin and EGCG, and curcumin possess strong cytotoxic activity and high binding affinity toward cancer-related target macromolecules. SAR analysis demonstrates that key structural features, including the number and position of free phenolic hydroxyl groups, the presence of gallate ester groups, and conjugated diketone systems, play a crucial role in determining ligand receptor complex stability. These interactions are supported by hydrogen bonding, hydrophobic interactions, and favorable steric compatibility within receptor binding sites. Computational findings further suggest that structural optimization can enhance ligand selectivity and improve pharmacokinetic properties. This study concludes that natural phytochemical scaffolds have significant potential as lead compounds and provide a rational basis for Computer-Aided Drug Design in developing more potent, selective, multi-target, and safer breast anticancer therapies.

Cutaneous fungal infections caused by Malassezia furfur remain a significant health concern in tropical regions, highlighting the need for safe and effective alternative therapies. Moringa oleifera L. leaves are reported to contain various secondary metabolites, including flavonoids, tannins, and saponins, which exhibit potential antifungal activity. This study aimed to develop a topical suspension formulation of Moringa oleifera leaf extract and to evaluate the effect of varying extract concentrations on the physical characteristics of the preparation. An experimental method was employed using four formulations: F0 as the control, and F1, F2, and F3 containing 9%, 10%, and 11% extract, respectively. Evaluation parameters included organoleptic properties, homogeneity, pH, viscosity, and adhesiveness, as well as stability testing using a cycling test method for 12 days. The results demonstrated that all formulations exhibited good organoleptic characteristics, homogeneity, and viscosity. The pH values remained within the acceptable range for skin preparations, although slight fluctuations were observed during storage. Adhesiveness also showed variations across several testing cycles. Overall, the formulations met acceptable physical quality criteria and demonstrated potential for further development as topical suspensions; however, formulation optimization is still required to enhance stability during storage.

Cutaneous fungal infections caused by Malassezia furfur remain a significant health concern in tropical regions, highlighting the need for safe and effective alternative therapies. Moringa oleifera L. leaves are reported to contain various secondary metabolites, including flavonoids, tannins, and saponins, which exhibit potential antifungal activity. This study aimed to develop a topical suspension formulation of Moringa oleifera leaf extract and to evaluate the effect of varying extract concentrations on the physical characteristics of the preparation. An experimental method was employed using four formulations: F0 as the control, and F1, F2, and F3 containing 9%, 10%, and 11% extract, respectively. Evaluation parameters included organoleptic properties, homogeneity, pH, viscosity, and adhesiveness, as well as stability testing using a cycling test method for 12 days. The results demonstrated that all formulations exhibited good organoleptic characteristics, homogeneity, and viscosity. The pH values remained within the acceptable range for skin preparations, although slight fluctuations were observed during storage. Adhesiveness also showed variations across several testing cycles. Overall, the formulations met acceptable physical quality criteria and demonstrated potential for further development as topical suspensions; however, formulation optimization is still required to enhance stability during storage.

Bank stock price prediction is an important topic in the application of information technology because stock price movements are dynamic, sequential, and influenced by historical market patterns. This study aims to predict Indonesian banking stock prices using the Long Short-Term Memory method and evaluate the effect of Bayesian Optimization on model performance. The data used in this study consists of daily historical stock data of BBCA, BBNI, BBRI, BBTN, and BMRI from May 4, 2020, to May 4, 2026, obtained from Yahoo Finance. The input features include opening price, highest price, lowest price, closing price, and trading volume, while the prediction target is the stock closing price. The results show that the baseline model produced MAPE values ranging from 1.892% to 3.147%. The best baseline performance was obtained on BBCA with an R² value of 0.933, followed by BBTN with an R² value of 0.902. After optimization, performance improvement occurred on BBTN, with MAPE decreasing from 3.147% to 2.482% and R² increasing from 0.902 to 0.935. For BMRI, MAPE decreased from 2.385% to 2.206%, and R² increased from 0.687 to 0.743. This study concludes that Long Short-Term Memory can be used to predict Indonesian banking stock prices, while Bayesian Optimization can selectively improve model performance depending on the characteristics of each stock dataset.

Poverty is a multidimensional problem that remains a major development challenge in Central Tapanuli Regency. Growth in Gross Regional Domestic Product (GRDP), increased realization of Foreign Direct Investment (PMA) and Domestic Direct Investment (PMDN), and zakat collection should be important instruments in reducing poverty rates. However, data from 2019–2023 shows a discrepancy, where the growth in these macroeconomic and social indicators is not accompanied by a decrease in the number of poor people. This study aims to analyze the influence of zakat, GRDP, PMA, and PMDN on poverty levels in Central Tapanuli Regency. The research method used is qualitative descriptive analysis with secondary data from BPS, BAZNAS, and BKPM as well as interviews with relevant parties. The results show that zakat does contribute to alleviating the burden on mustahik, but its role is still limited because the majority of distribution is consumptive. GRDP increased from Rp9.95 trillion in 2019 to Rp13.67 trillion in 2023, but the resulting economic growth is not yet inclusive. Foreign direct investment (FDI) tends to be oriented towards large capital with limited labor absorption, while domestic direct investment (PMDN) is closer to local needs but still less than optimal in poverty alleviation. Overall, the increase in zakat, GRDP, FDI, and PMDN has not been able to reduce the poverty rate, which actually increased from 376,474 people in 2019 to 489,760 people in 2023. Thus, a more inclusive development strategy, optimization of productive zakat, and investment policies that favor labor-intensive sectors and MSMEs are needed so that economic growth truly impacts poverty reduction.

Indonesia’s Free Nutritious Meal Program (MBG) requires an efficient and adaptive supply chain system to ensure timely distribution, cost efficiency, and adequate nutritional delivery for a large number of beneficiaries. However, conventional supply chain approaches are generally static and unable to respond effectively to dynamic demand, supply uncertainty, and logistical constraints. This study proposes a Multi-Objective Reinforcement Learning (MORL) model to optimize the MBG supply chain by simultaneously considering distribution cost, delivery timeliness, service level, nutritional adequacy, and food waste reduction. The model is developed using a simulation-based environment representing real-world supply chain conditions, including demand variability, transportation limitations, and kitchen capacity constraints. The results show that the proposed approach achieves cost reductions of 15–22%, improves delivery timeliness by 18–25%, maintains a service level above 90%, increases nutritional fulfillment by 12–18%, and reduces food waste by 10–15% compared to baseline methods. Sensitivity analysis further demonstrates the robustness of the model, with minimal performance degradation under disruption scenarios. These findings indicate that Reinforcement Learning provides a scalable and adaptive solution for optimizing large-scale public food distribution systems. The proposed model contributes both theoretically by integrating multi-objective optimization within an RL framework and practically by supporting data-driven decision-making for improving the effectiveness of the MBG program in Indonesia.

The improvement of village road infrastructure plays an important role in enhancing community mobility and supporting local socioeconomic development. This study aims to analyze changes in mobility conditions before and after road improvement, explore community perceptions and experiences, identify changes in activity and economic patterns, and examine factors that support or hinder mobility optimization in Trayeman Village, Slawi District, Tegal Regency. A mixed-method approach was employed involving 98 respondents selected from a population of 5,229 residents using the Slovin formula with a 10% margin of error, along with 15 key informants for qualitative data collection. Data were gathered through observations, in-depth interviews, documentation studies, and Likert-scale questionnaires. Quantitative data were analyzed descriptively, while qualitative data were examined using the Miles and Huberman interactive analysis model. The findings indicate that road improvement has significantly enhanced community mobility, with an overall mean score of 3.89, categorized as high. Access to services recorded the highest score (4.06), followed by economic mobility (4.00) and social mobility (3.99). Community perceptions were largely positive, viewing the improved road as a catalyst for development and improved accessibility. Road improvement also encouraged broader economic opportunities, lower transportation costs, business growth, and increased income. Supporting factors included private vehicle ownership and supportive village policies, whereas high fuel prices, limited transportation options, and uneven local road conditions remained major challenges. Further improvements in supporting infrastructure and transportation services are recommended to maximize mobility benefits.

The increasing demand for scalable and high-quality digital marketing content has exposed limitations in traditional manual copywriting processes, which are time-intensive and difficult to scale. This research proposes an adaptive AI-driven copywriting framework that integrates a full-stack web architecture with optimized prompt engineering strategies for automated content generation. The system is implemented using React.js for the frontend, Node.js with Express for backend services, and a GPT-based API for language generation. Unlike prior implementations, this research introduces a structured prompt optimization mechanism to enhance content relevance and consistency. Experimental evaluation was conducted using multiple datasets of marketing prompts, with comparisons against baseline GPT usage and manual copywriting. Quantitative results show that the proposed system achieves improvements in BLEU (+18.7%) and ROUGE-L (+21.3%) scores over baseline methods. Human evaluation involving 30 participants indicates a significant increase in perceived content quality, coherence, and persuasiveness (p < 0.05). System performance analysis demonstrates an average response time of 1.8–3.0 seconds and a GTmetrix performance score of 82%. The findings confirm that the proposed framework significantly enhances efficiency, scalability, and content quality, contributing to both applied AI systems and intelligent web-based content production.

People with disabilities still face various barriers in exercising their equal rights, including in accessing public facilities and tourism sites. This study aims to analyze the factors that support and hinder the implementation of Cimahi City Regulation No. 20 of 2018 on the Protection and Fulfillment of the Rights of Persons with Disabilities, particularly at tourist attractions in Cimahi City. The approach used is qualitative, employing a case study method to explore the phenomenon of policy implementation in depth using various data sources. The research findings indicate that supporting factors include strong legal legitimacy, the local government’s political commitment, coordination across Local Government Agencies (LGAs), the use of social media for outreach, and the involvement of the disability community in the planning process. Meanwhile, inhibiting factors include budget constraints, suboptimal accessibility facilities, a weak database system for disability needs, uneven public outreach, low empathy among tourism managers, and weak enforcement of sanctions. This study implies the need for strengthened oversight, optimization of resources, massive legal education, and a shift in mindset to achieve substantive inclusivity.

The increase in medical disputes in Indonesia occurs along with the development of health services and increasing public legal awareness. Dispute resolution through litigation is often considered ineffective because it takes a long time, is expensive, and creates a confrontational relationship between medical personnel and patients. Therefore, mediation is present as an alternative dispute resolution that prioritizes deliberation, communication, and mutual agreement. This study aims to analyze the regulation, implementation, and effectiveness of mediation as an alternative resolution of medical disputes based on Supreme Court Regulation Number 1 of 2016 concerning Mediation Procedures in Court. The research method used is normative legal research with a statutory, conceptual, and case approach. Data were obtained through literature and document studies, then analyzed descriptively qualitatively. The results of the study indicate that mediation has a strong legal basis and provides various advantages, such as a faster resolution process, lower costs, maintaining confidentiality, and being able to maintain good relationships between medical personnel and patients. However, the implementation of mediation still faces obstacles such as low public understanding, limited mediators who have competence in the health sector, and a legal culture that still tends to be litigative. Therefore, optimization is needed through improving mediator competence, strengthening public legal awareness, and maximizing mediation implementation to achieve fair, effective, and humane dispute resolution.

The acceleration of digitalization in recent years has shaped a new socio-economic landscape in Indonesia. Access to app-based financial services and online entertainment has increased rapidly, but at the same time, illegal online lending and digital gambling, targeting adolescents, have flourished. This paper explores a Community Service intervention model that combines preventative digital literacy and artificial intelligence (AI) optimization within the La Tansa Islamic Boarding School in Palembang. The research was conducted using a qualitative approach with participant observation, in-depth interviews, focus group discussions, and written reflection analysis. Findings indicate that strengthening digital literacy based on critical awareness not only improves risk understanding but also deepens self-control integrated with religious values. AI training for educators also encourages changes in learning practices to be more adaptive and reflective of the dynamics of the digital era. This model offers a digital resilience approach that can be replicated in other educational institutions with similar characteristics.

The use of synthetic dyes in the textile industry has caused various environmental problems due to waste that is difficult to degrade and has the potential to contaminate water and soil. Therefore, the development of environmentally friendly natural dyes is an important alternative to be explored. One potential but underutilized source of natural dye is kirinyuh weed (Chromolaena odorata L.), which is known to contain natural pigment compounds such as tannins and flavonoids. This study aims to analyze the dyeing quality of primissima cotton fabric using kirinyuh leaf extract with variations of alum and ferrous mordants through a mordanting process. The research employed an experimental method, including hot extraction of kirinyuh leaves, dyeing of primissima cotton fabric, and mordanting using alum and ferrous salts. The evaluation of dyeing quality focused on washing color fastness and light color fastness tests based on standard textile testing methods. The results indicate that the type of mordant significantly affects the resulting color quality. Alum mordant produced relatively lighter colors with good washing fastness, while ferrous mordant resulted in darker shades with similarly good washing fastness. However, color fastness to light showed relatively lower values, particularly in fabrics treated with alum mordant. Based on these findings, it can be concluded that kirinyuh leaf extract has the potential to be used as a natural dye for primissima cotton fabric, although further optimization is required to improve color resistance to light exposure.

Sign language recognition is one of the areas of image recognition and image processing technology that is developing rapidly in human-computer interaction. This technology really helps the deaf and speech impaired in communicating with non-disabled people. This research aims to examine the optimization of an object tracking system in sign language using the Gaussian Mixture Model (GMM) and Kalman Filter by including the Region of Interest (ROI). The proposed system consists of three main components, namely hand detection, object extraction, and classification. Hand detection is done using the Kalman Filter to track hand movements accurately. Next, Region of Interest (ROI) features, such as shape, direction and movement features, are extracted from the detected part of the hand. These features are fed into a Gaussian Mixture Model (GMM) classifier, which can recognize sign language based on the extracted features. With the combination of GMM and Kalman Filter in this research, it can increase accuracy in object tracking, reduce interference from the background, and ensure the tracking focus remains on important objects. The dataset used is in the form os SIBI alphabet symbols, namely A-Z with the amount of data for each class, namely 620 images. Based on the research result, model testing using GMM, Kalman Filter and ROI produces higher accuracy of 99%, while model testing using GMM and ROI produces accuracy of 90%.

The rise of social media has created a digital public sphere that enables users to express their opinions on social and political issues openly and in real-time. One of the most discussed topics on social media platform X is the trending hashtag #IndonesiaGelap, which reflects public concern and criticism regarding various governmental and societal conditions. This study aims to conduct sentiment analysis on tweets containing the hashtag to determine the overall sentiment trend among users. The method employed in this research is the Naive Bayes classification algorithm, known for its simplicity and effectiveness in text classification. To enhance the model’s performance, Particle Swarm Optimization (PSO) is applied to optimize feature selection and parameter tuning. The dataset consists of public tweets collected via the Twitter API, followed by preprocessing, feature extraction using TF-IDF, and sentiment classification into three categories: positive, negative, and neutral. The results indicate that the integration of PSO significantly improves the classification accuracy of the Naive Bayes model compared to the baseline. The majority of tweets related to #IndonesiaGelap exhibit a negative sentiment, indicating widespread public dissatisfaction and criticism. This research is expected to contribute to a better understanding of public perception and serve as valuable input for stakeholders in addressing social issues in the digital age.

This research aims to analyze the effectiveness of the Trans Banten Program in supporting the development of public transportation in Serang City. The research uses qualitative methods with a case study approach. Data was collected through observation, interviews, documentation and literature study, then analyzed interactively. The research results show that the Trans Banten Program is quite effective in increasing community accessibility and mobility, especially towards areas of education, government and public services. This service also provides economic benefits through reduced transportation costs and is supported by facilities that are relatively comfortable and safe for users. In addition, the increase in the number of passengers shows the public's acceptance and need for public transportation. However, the effectiveness of the program is not yet optimal because there are still obstacles in the form of fleet limitations, inaccurate operational schedules, passenger density during peak hours, and limited service information. Therefore, it is necessary to improve service quality, add fleets, and strengthen information systems to support the sustainability and optimization of the Trans Banten Program as public transportation in Serang City.

The development of social media has driven changes in digital marketing communication strategies, including within educational institutions such as Juara Academy. This study aims to analyze digital marketing communication strategies through Instagram in increasing student enrollment, identify gaps between social media performance and conversion rates, and formulate efforts to optimize these strategies. This research employs a descriptive qualitative approach using a case study method. Data were collected through interviews, observations, and documentation, and then analyzed using the interactive model of Miles and Huberman, as well as the AIDA framework (attention, interest, desire, action). The results indicate that digital marketing communication strategies have been systematically implemented through the utilization of Instagram features such as feed, story, reels, and live to capture attention, build interest, generate desire, and ultimately encourage enrollment actions. However, a gap was found between high engagement levels and suboptimal enrollment conversions, influenced by technical factors such as inconsistent content posting and a lack of integration across communication stages. Strategic optimization is therefore necessary through strengthening content consistency, enhancing more personalized interactions, and leveraging persuasive approaches based on audience experience. This study provides practical contributions to the management of digital marketing communication as well as theoretical contributions to the development of social media–based marketing communication studies.  

This community service program aims to implement digital marketing management strategies to optimize community based digital businesses among young generations in Jayapura City. The rapid development of digital technology has created significant opportunities for youth to develop digital businesses. however, limitations in marketing knowledge, content creation skills, and the use of digital platforms remain major challenges. This program was conducted through training, mentoring, and practical workshops focusing on digital marketing management, including market segmentation, branding strategy, social media marketing, content planning, digital advertising, and evaluation using digital analytics. The participants consisted of young entrepreneurs and youth communities who are actively involved in small scale digital business activities. The results of the program indicate an improvement in participant’s understanding and skills in managing digital marketing strategies, particularly in building brand identity, optimizing social media engagement, and designing digital promotional content. In addition, participants were able to develop structured digital marketing plans and apply them to their business activities. This program contributes to strengthening youth capacity in Jayapura City to compete in the digital economy through sustainable community based business development.

This Community Service (PKM) activity is motivated by the low capacity of MSME actors and start-up business people in utilizing social media and Artificial Intelligence (AI) technology as a means of digital promotion. These limitations include low digital literacy, limited capabilities for visual content production, weak marketing communication strategies, and inconsistent business social media management. This study aims to improve participants' understanding and skills in optimizing social media and utilizing AI to support business promotion. The approach used is a qualitative descriptive approach, with observation, interviews, documentation, direct practice, discussion, mentoring, and training evaluation. The activity was held in Petukangan, Pesanggrahan District, South Jakarta, with participants consisting of MSME actors, students, and the general public. The results showed that 85% of participants experienced an increase in understanding of digital marketing strategies.In comparison, 78% were able to create promotional content independently with the help of AI, especially through Bing Image Creator and ChatGPT. These findings show that integrating social media and AI can increase creativity, improve content production efficiency, and enhance the visual appeal of MSME promotion. The novelty of this activity lies in integrating AI-based visual design training with strengthening the digital entrepreneurship mindset. Thus, this training model can be an applicable, adaptive, and relevant MSME empowerment strategy for digital economy transformation.

This study aims to analyze the implementation of the madrasah quality paradigm from the perspective of the Qur'an at MIS Al-Hunafa Palangka Raya. The study used a qualitative field research approach with observation, interview, and documentation techniques. Data analysis used the Miles and Huberman model. The results of the study indicate that: 1) the quality paradigm is understood not only to be oriented towards academics, but also to the formation of character and spiritual values ​​of students; 2) the implementation of the quality paradigm is carried out through the integration of Qur'anic values ​​such as amanah, itqan, ihsan, discipline, and responsibility in madrasah culture, learning, and religious activities; and 3) the implementation of quality still faces obstacles in the form of limited infrastructure, variations in teacher competencies, and technological demands, so that the madrasah carries out strengthening through teacher training, facility optimization, and the development of a quality culture based on Qur'anic values. Thus, the implementation of the quality paradigm at MIS Al-Hunafa has been running quite well and is carried out sustainably.