Publication Search

The development of digital services by BPJS Ketenagakerjaan through the JMO (Jamsostek Mobile) application has triggered a surge in large-scale and unstructured user reviews on the Google Play Store, thereby complicating manual analysis and conventional sentiment analysis in accurately identifying specific issues. This research aims to implement the Aspect-Based Sentiment Analysis (ABSA) method to granularly evaluate JMO application reviews based on specific aspects, while simultaneously addressing class imbalance and computational efficiency issues. The proposed method combines the pretrained IndoBERT model as a contextual feature extractor, the SMOTE technique to balance the training data, and an artificial neural network (Neural Network) as the classification layer without performing full fine-tuning. The dataset used consists of 90,268 unique reviews categorized into five main aspects through keyword matching, namely General Satisfaction/Complaints, Performance & Stability, Service & Support, Feature Quality, and UI/UX, with initial lexicon-based labeling using the InSet Lexicon. The research results indicate that the proposed model successfully achieves highly optimal performance with an accuracy rate of 91.81% and a weighted F1-score of 92%. Furthermore, the implementation of SMOTE proved effective in enhancing model reliability on the minority class (negative sentiment), achieving an F1-score of 89%. The implications of this research contribute an accurate and efficient aspect-based sentiment analysis framework for developers, and serve as a strategic evaluation tool for BPJS Ketenagakerjaan in mapping specific user complaints to accelerate continuous improvements in the performance, stability, and service quality of the JMO application.

Agentic artificial intelligence (AI) systems capable of autonomous goal-directed behavior, multi-step planning, tool use, multi-agent coordination, and iterative self-correction represent a transition from passive clinical AI tools toward systems that can participate in complex healthcare workflows. However, empirical evidence remains fragmented across clinical decision support, patient monitoring, and administrative applications, and no systematic synthesis has evaluated which agentic principles have been technically demonstrated and which have accumulated sufficient evidence to support responsible clinical deployment. We conducted a PRISMA-informed systematic review of peer-reviewed empirical studies published between January 2025 and April 2026. Searches across five bibliographic databases and Google Scholar, supplemented by citation tracking, identified 443 unique records for screening, of which 25 met the predefined PICOS and quality appraisal criteria. Evidence was synthesized using an evidence-informed seven-principle framework derived from the integration of agentic AI, clinical AI, and healthcare governance literature. This framework provides a structured lens for examining how agentic principles are evaluated individually and in combination, enabling a deployment-readiness perspective that extends beyond capability-focused assessments alone. The evidence base was concentrated on technical capability principles, whereas human oversight, safety, compliance, and equity-related evaluation received comparatively limited attention. Most studies remained at the laboratory, benchmark, or proof-of-concept stage, and none reported demographic-stratified performance outcomes. Overall, the findings suggest a structural asymmetry in agentic healthcare AI: empirical research is advancing agentic capabilities more rapidly than it is generating evidence for the oversight, safety, equity, and governance mechanisms required for responsible clinical translation.

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.

This study examines human resource competence in improving financial management at the Public Service Agency of Gorontalo State University, a public higher education institution that operates under a flexible financial management model while remaining accountable for public funds. The main problem addressed is how financial management personnel translate regulatory knowledge, technical skills, and professional attitudes into efficient, effective, and accountable financial governance. This study aims to analyze the competence of financial management personnel and explain its contribution to strengthening institutional financial management. A qualitative descriptive approach was employed through interviews, observation, and document analysis involving bureau leaders, financial work team officials, treasurers, and financial managers across relevant work units. The findings show that knowledge competence is reflected in personnel understanding of regulations, policies, financial systems, budgeting procedures, reporting requirements, and the linkage between budget and institutional performance. Skills competence is demonstrated through financial administration, transaction recording, document verification, use of financial information systems, reconciliation, reporting, and preparation of accountability documents. Attitudinal competence appears in professionalism, compliance, integrity, prudence, responsibility, and openness to evaluation and audit. Financial management has been directed toward performance-based planning, expenditure control, budget realization monitoring, reporting, supervision, and audit follow-up. However, challenges remain in regulatory adaptation, system integration, data quality, document timeliness, account-code accuracy, inter-unit coordination, and consistency of audit follow-up. The study concludes that strengthening human resource competence is essential for improving financial management that is efficient, effective, accountable, and performance-oriented in public university financial governance.

Discrimination in the workplace is a problem that can hinder the creation of a fair and inclusive work environment. Discrimination can take the form of differential treatment of employees based on gender, race, religion, disability, sexual orientation, or other factors unrelated to individual performance and competence. To address this issue, the law plays a crucial role in providing employee protection to ensure equality and non-discrimination in the workplace. Various legal instruments, both national and international, regulate employee protection from discrimination, including the Employment Law, the Human Rights Law, and conventions issued by the International Labour Organization (ILO). However, the effective implementation of these regulations remains a challenge, particularly in terms of implementation, enforcement, and employee awareness of their rights. This study aims to analyze the role of law in protecting employees from discrimination in the workplace by examining applicable regulations and the challenges in their implementation. The research method used is a normative juridical method, which focuses on the study of relevant laws and legal principles. The analysis is conducted on national legal provisions and international legal instruments as references for protecting workers from discrimination. Furthermore, this study identifies barriers to legal implementation and offers recommendations to improve the effectiveness of legal protection for employees. This analysis is expected to provide insight into the urgency of regulatory reform and strengthening so that the law can play an optimal role in creating a fairer and more discrimination-free work environment.

The Free Nutritional Meal Program (MBG) is a government policy that is widely discussed by the public through social media, especially TikTok. Various comments that have emerged indicate differences in public opinion towards the program, so an analysis is needed to determine the tendency of public sentiment. This study aims to analyze TikTok user sentiment towards the Free Nutritional Meal Program using the Naive Bayes method. The research method is carried out through several steps, namely collecting TikTok comment data, preprocessing text, labeling sentiment data into positive, negative, and neutral, feature transformation using TF-IDF, and classification using the Naive Bayes algorithm. Based on the analysis of 500 comment data, the results show that positive sentiment dominates public opinion by 42% (210 data), followed by negative sentiment by 36% (180 data), and neutral sentiment by 22% (110 data). Testing the classification model using Naive Bayes produces excellent performance with an accuracy rate of 86%, precision of 84%, recall of 85%, and F1-score of 84%. The conclusion of this study shows that the Naive Bayes method is effective as an approach in social media sentiment analysis to map public responses to government policies.

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.

Batik motif classification has attracted growing attention in visual computing due to its role in cultural heritage preservation, textile informatics, museum documentation, and automated cataloging. Although many studies report high classification accuracy, robustness under real-world acquisition conditions remains insufficiently understood. Batik images are frequently affected by illumination variation, blur, folds, watermark overlays, wearable deformation, scale inconsistency, and background clutter, creating challenges that extend beyond conventional image-noise assumptions. Existing studies largely focus on improving classification performance, while the interactions among acquisition variability, feature representation, evaluation practice, and deployment constraints remain fragmented. This systematic literature review addresses this gap by synthesizing batik classification research through a robustness-aware perspective. Using query expansion, backward and forward citation chaining, relevance screening, and thematic coding, 116 candidate records were identified, resulting in 50 highly relevant studies for detailed analysis. The review reveals that robustness is shaped less by denoising alone than by the combined effects of acquisition conditions, representation design, evaluation realism, and deployment context. Handcrafted descriptors remain competitive for small datasets and structured motifs due to their data efficiency and interpretability, whereas deep learning models achieve the highest reported accuracy when supported by sufficient data diversity and realistic augmentation. Hybrid representations emerge as the most consistently balanced approach, combining local texture stability with higher-level abstraction across heterogeneous acquisition settings. The review further identifies recurring robustness failure patterns, including background dependency, illumination instability, motif-scale inconsistency, wearable deformation, and source-shift vulnerability. Based on these findings, a robustness-oriented research agenda is proposed, emphasizing cross-acquisition evaluation, representation-stability analysis, batik-specific robustness benchmarks, acquisition-aware augmentation, and deployable lightweight or hybrid architectures. The study contributes a domain-specific synthesis that reframes batik motif classification from an accuracy-centric task toward a robustness-aware visual recognition problem.

Python programming has become a fundamental competence in the digital era, yet students often struggle to transform algorithmic logic into functional code. This gap between conceptual understanding and practical implementation skills requires a thorough investigation into learning challenges within the Industrial Informatics Engineering Technology (TRIN) program at Politeknik Manufaktur Bandung. Grounded in Bloom's Revised Taxonomy and Cognitive Load Theory, this descriptive quantitative study utilized a Likert-scale questionnaire and an objective comprehension test administered to 87 third-year students. Data were analyzed using descriptive statistics to map performance across three aspects: conceptual understanding, syntactic comprehension, and implementation ability. Results indicate the conceptual aspect achieved the highest average of 4.15, followed by syntax at 3.56 and implementation at 3.54, with objective test accuracy rates of 76.09%, 65.52%, and 67.36%, respectively. Major obstacles identified include difficulties with looping, debugging, and comparison operators. Therefore, enhanced structured practice and Project-Based Learning approaches are recommended to strengthen students' implementation competencies.

This study examines the role of the creative economy in enhancing sustainable tourism potential in Pampang Cultural Village, Samarinda, Indonesia, and identifies strategic priorities for its development. Pampang Cultural Village is recognized for its rich cultural heritage preserved by the Dayak Kenyah community, which serves as a valuable tourism asset. Despite this potential, the contribution of creative economy activities to tourism development has not been fully optimized. Therefore, a comprehensive analysis is needed to formulate effective development strategies. This research employs a mixed qualitative–quantitative approach using the SWOT-AHP method. SWOT analysis is utilized to identify internal strengths and weaknesses as well as external opportunities and threats affecting creative economy development. Subsequently, the Analytic Hierarchy Process (AHP) is applied to determine the priority level of each strategic factor and alternative strategy. The findings indicate that the village possesses significant creative economy potential in cultural performances, handicrafts, culinary products, traditional fashion, and cultural souvenirs. Cultural authenticity emerges as the primary strength, while limited innovation capacity and digital marketing skills remain the major weaknesses. The results further reveal that the highest-priority strategy is the development of innovative creative tourism products rooted in Dayak Kenyah cultural heritage while preserving cultural authenticity. This strategy should be supported by digital promotion, human resource capacity building, stakeholder collaboration, and improvements in tourism infrastructure to achieve sustainable tourism development.

Medical students need to take the Competence test as Doctor Profession Program (UKMPPD) to be passed and earned their medical profession. Exams are one of the most common stressors experienced by medical students and the body will respond to these stressors in the form of feelings of depression or anxiety. Anxiety and depression itself can affect student performance during exam preparation. The purpose of this study was to determine the relationship between anxiety and depression on the preparation of the Professional Doctoral Program Competency Test (UKMPPD) for students of the Faculty of Medicine, University of Pattimura. The method used is quantitative analysis with cross sectional research design and total sampling technique. The sample is all the student population of the Faculty of Medicine, University of Pattimura who meet the inclusion and exclusion criteria, totaling 90 respondents. The results showed that the majority of respondents experienced mild anxiety symptoms, which is 37% during the CBT exam and 34% during the OSCE exam. The majority of respondents also did not experience depression, which is 68% during the CBT and 70% during the OSCE and the majority had an optimal level of preparation for UKMPPD, which is 63% during the CBT and 64% during the OSCE. Data analysis using the chi square test showed a significant relationship between anxiety symptom and the level of preparation for UKMPPD both CBT (p=0.030) and OSCE (p=0.012). There was no significant relationship between depression symptom and the level of preparation for UKMPPD for both CBT (p=0.123) and OSCE (p=0.07). It can be concluded that there is a significant relationship between anxiety and preparation for UKMPPD, but there is no significant relationship between depression and preparation for UKMPPD.

TPS 3R is a community-based waste management approach aimed at reducing waste generation through reduce, reuse, and recycle activities. However, the operational sustainability of TPS 3R still faces several challenges,including  waste processing effectiveness, and community participation. This study aims to analyze the community’s Willingness to Pay (WTP) for waste management fees and to formulate priority recommendations for improving the performance of TPS 3R Banjar Sugihan using the Quality Function Deployment (QFD) method. The research was conducted in Banjar Sugihan Village, Tandes District, Surabaya City, involving 563 household respondents. WTP analysis was carried out using the bidding game method, while QFD analysis was conducted through the preparation of the House of Quality (HoQ). The results showed that the community’s WTP ranged from IDR 16,000 to IDR 19,000 per month. If applied to all 3,758 households in Banjar Sugihan Village, the potential retribution revenue is estimated to reach IDR 60,128,000–IDR 71,402,000 per month. The QFD analysis indicated that the priority strategies for improving TPS 3R performance include enhancing infrastructure facilities, optimizing waste sorting and processing, increasing waste processing capacity, strengthening human resource capacity, and implementing a WTP-based retribution system. Therefore, the implementation of a WTP-based retribution system and priority strategies derived from QFD analysis are expected to support the sustainability of waste management at TPS 3R Banjar Sugihan.

In the Education 4.0 era, schools face unprecedented challenges in managing teacher performance effectively. Traditional top-down performance management systems have proven inadequate for fostering continuous professional growth, intrinsic motivation, and adaptive teaching competencies required in increasingly technology-driven learning environments. This study aims to develop, validate, and assess the effectiveness of a School-Based Performance Management (SBPM) model designed to enhance teacher effectiveness across cognitive, affective, and pedagogical dimensions. Using a Research and Development (R&D) design following the ADDIE framework, the study was conducted in three phases: needs analysis, model development, and model validation. Participants included 12 school principals, 186 teachers, and 8 education experts from 24 public secondary schools in [Province, Country]. Validation by experts yielded a content validity index (CVI) of 0.91. Implementation resulted in statistically significant improvements in teacher effectiveness scores (t = 8.74, p < 0.001, Cohen's d = 1.23), digital pedagogy competency (mean increase = 22.4%), and student learning outcomes (mean improvement = 17.8%). The SBPM model provides a contextually responsive, evidence-based framework that empowers schools to manage teacher performance collaboratively, fostering professional accountability and sustainable instructional quality in Education 4.0.

Rapid population growth in Sidoarjo Regency has triggered massive land-use changes, resulting in increased surface runoff and reduced performance of the drainage system. This study aims to evaluate the hydraulic capacity of drainage channels in the Pondok Sidokare Indah Housing area against design flood discharges with return periods of 2, 5, and 10 years. The method used is a descriptive quantitative approach, involving hydrological analysis using maximum daily rainfall data from 2015–2025 and hydraulic modeling of the existing channel along 350 meters. The frequency analysis results indicate that the Log Pearson Type III distribution is the most suitable method based on statistical parameters and the Smirnov-Kolmogorov goodness-of-fit test. The calculation of design flood discharge using the rational method yields values of 0.749 m³/s (2-year), 1.003 m³/s (5-year), and 1.164 m³/s (10-year). Meanwhile, the maximum capacity of the existing channel ranges only between 0.534 m³/s and 0.733 m³/s. The comparison between hydrological load and channel capacity shows that all observation points (Sta 0+000 to Sta 0+350) are in overflow condition, even for the lowest return period flood discharge. This condition confirms that the current channel dimensions are no longer adequate and require normalization to mitigate annual flooding in the area.

Construction project control requires careful planning to avoid delays and cost overruns. This study aims to evaluate the performance of the CBT Laboratory Building construction project at the Faculty of Medicine, Wahid Hasyim University, Semarang, in terms of cost and time using the Earned Value Analysis (EVA) method, and to analyze acceleration alternatives using the crashing method. The method used is a quantitative approach with a case study, based on project data such as the Budget Plan (RAB), S-curve, and work progress reports. The analysis was carried out using the Earned Value Analysis (EVA) method to assess project performance in terms of cost and time through indicators such as CV, SV, CPI, and SPI. Furthermore, an acceleration analysis was conducted using the crashing method to determine the optimal alternative for overcoming project delays by considering time and cost efficiency. The results of the Earned Value Analysis indicate that the project experienced delays and cost overruns, as shown by SPI < 1 and CPI < 1. The estimated project completion time increased to 227 days, longer than the initial plan of 217 days, while the estimated final cost reached RP 5,451,241,064.85, exceeding the initial budget. Acceleration efforts using the crashing method show that adding adding labor is more efficient than working hours (overtime), resulting in a project duration of 212 days with lower costs of RP 5,658,221,364.08. Thus, the Earned Value Analysis method is effective for evaluating project performance, while crashing can be a solution for acceleration by considering time and cost efficiency.

Poor soil quality and inefficient nutrient utilization remain major constraints in tomato (Solanum lycopersicum L.) cultivation, often resulting in reduced plant growth, lower productivity, and suboptimal fruit quality. This study aims to evaluate the growth performance of rose tomatoes through the integration of open-field cultivation and hydroponic planting systems. The research methods included land preparation using the Pythagorean technique to ensure proper spacing and planting layout, the application of organic fertilizers to improve soil fertility, and the transition to a hydroponic drip irrigation system using rockwool as the growing medium during weeks 1 to 7 of plant development. Plant growth parameters, including plant height, leaf development, stem vigor, and overall plant health, were observed throughout the cultivation period. The results indicated that the combination of appropriate soil management practices and precise nutrient delivery in the hydroponic system significantly enhanced nutrient uptake and supported healthy plant growth. Furthermore, the drip irrigation system helped optimize water use efficiency while reducing nutrient losses. This integrated cultivation approach demonstrated its potential as an effective solution for maintaining crop quality and productivity, particularly in areas with limited land resources. Overall, the findings suggest that integrated management practices can provide a practical, efficient, and sustainable strategy for farmers seeking to improve horticultural production in modern agricultural systems.

Education is the main driver of a nation's progress, which is not only related to intellectual intelligence, but also the formation of student character. Unfortunately, currently the world of education is facing serious problems, such as declining interest in learning and increasingly complex external factors in the digital era. This study aims to explore in depth the impact of learning motivation on student academic achievement, identifying obstacles that hinder this achievement. The method used in the current study is a literature study with a qualitative approach. Data were collected through theoretical studies and analysis of various relevant scientific sources, including research journals and textbooks, which were then combined to develop a comprehensive argument. The findings of this study indicate that learning motivation is the main factor that encourages student persistence and consistency in achieving the best learning. Academic success is defined as the result of a complex interaction between internal and external factors, where the existence of a supporting ecosystem such as good learning facilities, a supportive environment, and effective communication between lecturers and students plays a very important role. On the other hand, phenomena such as digital fatigue, low independence, and excessive workload are identified as significant barriers that can reduce academic performance. This analysis concludes that there is synergy in strengthening psychological aspects and creating an adaptable learning environment to maintain the stability of academic achievement amidst global demands.

The consistent decline in cucumber production in Indonesia indicates limitations in conventional cultivation systems, particularly due to land scarcity and inefficient resource management. This condition has encouraged the development of hydroponic systems as a more controlled and productive cultivation alternative. This study aims to critically analyze the key factors determining the success of hydroponic cucumber cultivation and to identify the most influential management aspects in improving yield. The method employed is a literature review, examining various recent studies related to hydroponic systems, nutrient management, growing media, and environmental factors. The results show that the advantages of hydroponics lie not only in land and water efficiency but also in the ability to precisely control growth variables. However, optimal productivity highly depends on the proper integration of nutrient management, particularly the regulation of pH, electrical conductivity (EC), and nutrient balance, as well as the control of environmental factors such as temperature, humidity, and light intensity. Inaccuracy in a single component can significantly reduce plant performance, even when other factors are optimal. Therefore, an integrated approach combining nutrient and environmental management simultaneously is essential to enhance hydroponic cucumber productivity. This study confirms that hydroponics has strong potential as a strategic solution to support sustainable agriculture amid land limitations in Indonesia.

This study examines the performance of student administrative services in the Society 5.0 era at the Bureau of Academic Affairs, Student Affairs, and Planning of Gorontalo State University, focusing on scholarship services. It addresses the need for accessible, responsive, transparent, inclusive, and student-centered administrative services in higher education. A descriptive qualitative approach was employed through in-depth interviews, participatory observation, and document analysis. Informants included bureau leaders, scholarship and financial aid administrators, operational officers, and students receiving various scholarship schemes. Thematic analysis was conducted through data reduction, data display, and conclusion drawing, while validity was ensured through source triangulation, method triangulation, and member checking. The findings show that accessibility has improved through service counters, websites, social media, WhatsApp, online forms, SIMPEL BAKP, and coordination with faculties and study programs. However, information remains fragmented, digital standard operating procedures are not uniform, and disability-inclusive access is limited. Responsiveness is supported by direct and digital communication, but it lacks a dedicated complaint system, automatic notifications, selection-status dashboards, and real-time disbursement tracking. Service quality is perceived as fairly good because students experience professional, friendly, and fair treatment. Nevertheless, service documentation, procedural standardization, staff training, and humanistic technology integration need strengthening. The study concludes that scholarship administration should be transformed into a centralized, inclusive, responsive, transparent, and student-centered digital-humanistic service model.

Abstract: Detecting chili leaf diseases remains challenging due to the non-uniform manifestation of symptoms, local discoloration, small lesion regions, and visual similarity between disease patterns and natural leaf background variations. Although YOLO-based detectors provide favorable computational efficiency, lightweight variants often struggle to distinguish subtle lesion characteristics, while conventional attention mechanisms such as CBAM primarily rely on global feature aggregation and may overlook regional activation variability. To address these limitations, this study proposes a YOLOv9s-based detection framework integrated with a Region-Dispersion Channel Spatial Attention (RDCSA) module. The proposed module incorporates regional dispersion statistics, namely mean, standard deviation, and range, as channel descriptors to capture inter-region feature variability before applying spatial attention refinement. Experiments were conducted on the COLD dataset containing 532 original images from five chili leaf condition categories using a split-before-augmentation protocol to ensure objective evaluation. RDCSA was integrated at the P5 feature level and evaluated through attention placement analysis, component-wise ablation, sensitivity analysis, stability assessment, and comparison with modern attention mechanisms. The proposed YOLOv9s + RDCSA model achieved an mAP@50 of 0.894, mAP@50–95 of 0.773, precision of 0.858, recall of 0.861, and an F1-score of 0.859 with only a marginal increase in model parameters. The results suggest that regional dispersion-based attention improves feature discrimination while preserving computational efficiency, particularly for disease symptoms characterized by heterogeneous spatial patterns. Nevertheless, performance remains influenced by visually ambiguous symptom categories, indicating that further validation across multiple datasets and field conditions is required. Overall, the proposed RDCSA module enhances detection capability without substantially increasing computational overhead, making it a promising attention mechanism for lightweight plant disease detection systems.