Publication Search

Tomato crop yields face significant threats from plant diseases, with existing deep learning solutions often computationally prohibitive for resource-constrained agricultural settings; to address this gap, we propose Efficient Disease Attention Network (EDANet), a novel lightweight architecture combining depthwise separable convolutions with hybrid attention mechanisms for efficient Tomato disease recognition. Our approach integrates channel and spatial attention within hierarchical blocks to prioritize symptomatic regions while utilizing depthwise decomposition to reduce parameters to only 104,043 (multiple times smaller than MobileNet and EfficientNet). Evaluated on ten tomato disease classes from PlantVillage, EDANet achieves 97.32% accuracy and exceptional (~1.00) micro-AUC, with perfect recognition of Mosaic virus (100% F1-score) and robust performance on challenging cases like Early blight (93.2% F1) and Target Spot (93.6% F1). The architecture processes 128×128 RGB images in ~23ms on standard CPUs, enabling real-time field diagnostics without GPU dependencies. This work bridges laboratory AI and practical farm deployment by optimizing the accuracy-efficiency tradeoff, providing farmers with an accessible tool for early disease intervention in resource-limited environments.

Chronic kidney disease (CKD) is a progressive and irreversible decline in kidney function that, if left untreated, can lead to serious complications. Ultrasonography (USG) is a widely used imaging modality for detecting CKD, yet its interpretation remains highly dependent on the radiologist’s expertise. This study aims to develop a CKD detection system using a convolutional neural network (CNN) on kidney ultrasound images based on estimated glomerular filtration rate (eGFR), and to evaluate the system’s performance. This research employed a research and development (R&D) approach with an experimental design. The dataset consisted of kidney ultrasound images from CKD and non-CKD patients with corresponding eGFR values. The methodology included image preprocessing, CNN model training, and accuracy evaluation using classification metrics. The results demonstrated that the developed CNN model achieved a total accuracy of 97% on internal test data and 95.8% on external validation. The model’s sensitivity reached 100% for the normal category, 91.67% for CKD stage 4, and 90% for CKD stage 5. Specificity exceeded 96% across all categories, with high precision and F1-scores above 94% for all classes. This system has proven to be effective as a diagnostic support tool for automatically detecting CKD through kidney ultrasound imaging. Its advantages lie not only in accurately classifying CKD from USG images but also in correlating the classification results with patients' eGFR values. This provides more precise clinical information and supports appropriate CKD staging and management planning.

This study examines the application of Educational Data Mining (EDM) to predict the academic per-formance of first-year students at the Catholic University of Bukavu and the Higher Institute of Edu-cation (ISP) in the Democratic Republic of Congo. The primary objective is to develop a model that can identify at-risk students early, providing the university with a tool to enhance student support and academic guidance. To address the challenges posed by data imbalance (where successful cases outnumber failures), the study adopts a hybrid methodological approach. First, the SMOTE algorithm was applied to balance the dataset. Then, a stacking classification model was developed to combine the predictive power of multiple algorithms. The variables used for prediction include the National Exam score (PEx), the secondary school track (Humanities), and the type of prior institution (public, private, or religious-affiliated schools), as well as age and sex. The results demonstrate that this approach is highly effective. The model is not only capable of predicting success or failure but also of forecasting students' performance levels (e.g., honors or distinctions). Moreover, the use of the Apriori association rule mining algorithm allowed the identification of faculty-specific success profiles, transforming prediction into an interpretable decision-support tool. This research makes several significant contributions. Practically, it provides the University of Bukavu with a tool for student orientation and early risk detection. Methodologically, it illustrates the effectiveness of a combined approach to EDM in an African context. However, the study acknowledges certain limitations, including the non-public nature of the data and the geographical specificity of the sample. It therefore proposes avenues for future research, such as the integration of Explainable AI (XAI) techniques for more refined and transparent analysis of the results.

This study aims to identify words that contain lexical ambiguity, explore their meanings, and classify them in the children’s book Amelia Bedelia by Herman Parish. The data were collected from the Amelia Bedelia Chapter Books series, which consists of twelve books published between 2013 and 2018. The analysis identified a total of 58 words with lexical ambiguity, including 46 instances of polysemy and 12 of homonymy. The classification of polysemy and homonymy is based on the semantic relatedness or unrelatedness between the multiple meanings of each word. The Amelia Bedelia stories effectively illustrate how children perceive and react to lexical ambiguity through the main character, Amelia, who is known for her literal interpretations of figurative language. This literal understanding often leads to humorous situations. Lexical ambiguity in these books functions not only as a source of entertainment but also as a valuable linguistic learning tool. Therefore, the stories contribute to language development in early childhood readers.

The Family Hope Program (PKH) is a conditional social assistance program provided by the government to improve the quality of life of underprivileged families through support in the education, health, and social welfare sectors. In its implementation, the process of determining PKH candidate recipients at the West Sumba Regency Social Service often experiences obstacles, especially with regard to objectivity, accuracy of targets, and limitations in complex data management. Thus, a decision support system (SPK) is needed that can assist the agency in selecting prospective recipients more effectively, efficiently, and on target. This study proposes the application of the Naive Bayes method in the development of SPK to determine PKH recipients. The Naive Bayes method was chosen because of its ability to classify data based on probability, and it can handle large volumes of data with a good degree of accuracy. The criteria applied in the classification include the level of household income, the number of members covered, the state of residence, the education of children, and the health of family members. The research process includes needs analysis, system design, data collection, application of Naive Bayes algorithms, and system testing. The findings of the study show that SPK based on Naive Bayes can provide recommendations for PKH recipients with better accuracy compared to manual methods. In addition, the system is able to improve transparency, fairness, and speed in the recipient selection procedure. With this system, it is hoped that the distribution of PKH in West Sumba Regency can be more orderly, balanced, and on target in accordance with the goals of government programs.

The study used a pseudo-experimental design with a posttest-only control group design pattern. The population included the entire class VII, and the sample was randomly selected of two equivalent parallel classes, totaling 22 students each. The experimental class received the discovery learning treatment, while the control class used conventional methods. The research instrument is in the form of learning outcome tests in the form of objectives and essays that have been tested for validity and reliability. Data were analyzed through normality test, homogeneity test, and t-test of two independent samples at a significance level of 0.05. The results showed significant differences between the two groups. The average posttest score of the experimental class was 80.68 higher than the control class of 72.72. The analysis of the t-test yielded a t_hitung value of 2.74 greater than t_tabel 2.01, which means that the null hypothesis was rejected. The frequency distribution in the experimental class also showed a concentration of scores in the high category, while the control class was more dominant in the medium category. These findings confirm that discovery learning not only improves cognitive learning outcomes, but also encourages students' active engagement, motivation, and critical thinking skills. This study concludes that the implementation of discovery learning is effective in improving science learning outcomes in the classification of living things. Implicitly, teachers are advised to integrate this approach as an alternative to student-centered science learning strategies.

This research presents the design and testing of an automatic color detection system using TCS3200 color sensor integrated with Arduino Uno microcontroller. The system was developed and tested using Wokwi virtual simulation platform before physical implementation. The TCS3200 sensor converts RGB light intensity reflected from objects into frequency signals, which are processed by Arduino Uno to classify colors into red, green, and blue categories. The system incorporates audio feedback using DFPlayer Mini module to provide sound notifications for detected colors. Testing results show that the system can accurately detect and classify primary colors with frequency-based thresholds: red (R<48 &R>37 & G<95 & G>85), blue (G<75 & G>65 & B<33 & B>23), and green (R<55 & R>40 & B<25 & B>5). The simulation validation demonstrates stable performance with consistent color recognition capabilities, making it suitable for industrial sorting applications and assistive technology for visually impaired individuals.

Criminal law based on genetic determinism was once rejected in modern criminal law systems because it was deemed contrary to the principle of individual responsibility. However, the rise of epigenetics and neurocriminology in contemporary legal practice indicates a reconstitution of the biological basis for attribution of criminal culpability. This study aims to analyze the extent to which developments in epigenetics reopen opportunities for the operation of biological approaches in criminal law, while also critiquing the conceptual dangers they pose to the principles of justice and moral responsibility. The research method uses a normative legal approach with a conceptual approach. The results show that epigenetics works as a tool for scientific validation of the formation of risk categories in criminal law, while simultaneously weakening the perpetrator's position as a moral subject. The criminal law structure that technocratically accepts biological arguments creates a new form of legal exclusion through medical classifications that are not open to ethical evaluation. In this situation, the law operates as an instrument of biological management of bodies deemed deviant. The position of neurocriminology in this case is no longer merely a tool, but rather the center of the configuration of biolegal power that defines responsibility based on predisposition, not will. Therefore, a new normative framework is needed that can uphold the principle of individual responsibility while rejecting the ethical reduction of biological diagnoses in the criminal law system.

This research examines directive speech acts in the “Burnout” playlist on the Satu Persen YouTube channel. The aim of this research is to analyze the classification of types, contextual meanings, functions, and intentions of directive speech acts in the “Burnout” playlist on the Satu Persen YouTube channel. The research method used is a descriptive qualitative methodological approach with a pragmatic theoretical approach. Analyzing directive speech acts through comprehensive examination based on relevant theories and presenting the findings descriptively. Directive speech acts in the “Burnout” playlist include directive requestives, directive questions, directive requirements, directive prohibitives, directive permissive, and directive advisories. The research is expected to contribute to enriching pragmatic studies and expanding public awareness of pragmatics in communication. Understanding the variations in directive speech acts patterns in YouTube social media and understanding their correlation with societal realities can serve as a foundation for the public to engage more intelligently and consciously in building effective, wise, and multidimensional communication.

The Emergency Severity Index (ESI) is a triage method that determines the escalation of treatment for patients based on the severity of their emergency condition. In order to improve the level of satisfaction of individuals receiving health services, it is necessary to provide quality services that are responsive to the expectations and needs of patients. This study aims to explore the relationship between the ESI (Emergency Severity Index) level and patient satisfaction in the Emergency Room of Pertamedika Ummi Rosnati Hospital. The research design applied is analytical with a cross-sectional method. From a total population of 500 patients in December, 51 respondents were selected using purposive sampling. The instruments used included the ESI (Emergency Severity Index) Questionnaire and CSQ-8 (Client Satisfaction Questionnaire), with univariate and bivariate data analysis using the Chi-Square test. The results showed that in the high ESI index category, most participants (37 people or 92.5%) expressed satisfaction with the services received, while 3 people (7.5%) were dissatisfied. In the moderate ESI group, 9 respondents (90.0%) reported satisfaction, while only 1 respondent (10.0%) reported dissatisfaction. Conversely, in the mild ESI level, there were no patients who were satisfied (0%), and one patient (20.0%) was dissatisfied. Through analysis using the Chi-Square statistical test, a ρ value of 0.009 (≤ 0.05) was obtained, indicating a significant relationship between ESI levels and patient satisfaction. Based on these findings, it can be concluded that there is a correlation between ESI classification and the level of satisfaction of service users in the Emergency Department. The researchers recommend that patients gain a better understanding of the ESI stages, and the Emergency Department is expected to provide education, such as putting up informative banners related to ESI (Emergency Severity Index) to increase patient understanding of the triage process in the Emergency Department.

Speech delay in children is a developmental issue commonly encountered in society, which can affect various aspects of a child's life, including communication, social interaction, and academic development. Early detection of speech delay is crucial for providing appropriate interventions to minimize its long-term impact on the child. This study aims to introduce the use of machine learning algorithms in detecting speech delay symptoms in children. Three machine learning algorithms applied in this study are Naïve Bayes, C4.5, and K-Nearest Neighbor (K-NN). These algorithms are used to classify speech delay symptoms based on health data, medical history, and environmental factors such as speaking habits and eating patterns. The outreach was conducted at Puskesmas Kota Rantauprapat with the involvement of parents and healthcare providers as participants. The experimental results showed that all three algorithms performed well in terms of accuracy, though with varying error rates. Naïve Bayes achieved relatively high accuracy but had a higher false positive rate compared to C4.5 and K-NN. C4.5 provided more stable results and was easier to interpret due to its decision tree structure. Meanwhile, K-NN performed better with data that had irregular distribution. This outreach is expected to assist both the community and healthcare providers in early detection of speech delay in children, providing a more efficient and affordable means for early intervention, which ultimately leads to better outcomes for children with speech delay.

Traffic violations are one of the serious problems frequently occurring in various regions, including Binjai City. Various types of violations, such as disobeying road signs and markings, incomplete vehicle documents, and violations that threaten the safety of drivers and other road users, continue to increase despite preventive and repressive efforts carried out by the authorities. This condition indicates that handling traffic violations cannot rely solely on field enforcement but also requires the support of technology capable of analyzing data more comprehensively. This study aims to predict the level of traffic violations by applying the Naïve Bayes method through data mining techniques. The dataset used consists of traffic violation records in 2023 from the Binjai City Police Department, with the main variables including violations of traffic signs and markings, document completeness, and safety-related violations. The Naïve Bayes method was selected because of its ability to perform classification with good accuracy, simplicity, and efficiency in processing large amounts of data. The implementation of this research is realized by developing a web-based application using Visual Studio Code as the development environment and MySQL as the database system. The results of this study are expected to provide structured information regarding traffic violation patterns, support authorities in making more effective decisions, and serve as an alternative solution in the prevention and handling of traffic violations in Binjai City.

Sleep is a fundamental human need that plays a crucial role in maintaining both physical and mental health. Poor sleep quality can trigger a variety of health problems, ranging from decreased concentration to an increased risk of chronic diseases. The complexity of factors influencing sleep quality—such as stress levels, heart rate, blood pressure, physical activity, and lifestyle—makes its assessment difficult through direct observation alone. Therefore, data mining approaches are increasingly utilized to identify relevant patterns in sleep-related data. This study aims to compare the performance of the C4.5 (Decision Tree) algorithm and the Naïve Bayes algorithm in predicting sleep quality using the Sleep Health and Lifestyle dataset, which contains information from 374 respondents. The research method applied is a quantitative comparative approach employing classification techniques with 10-fold cross-validation to ensure robust evaluation. Model performance is assessed using accuracy, precision, and recall metrics to provide a comprehensive understanding of the effectiveness of each algorithm. The findings indicate that the C4.5 algorithm achieves an accuracy of 96.26% and offers advantages in terms of interpretability through its decision tree visualization, enabling easier understanding of variable relationships. In contrast, the Naïve Bayes algorithm demonstrates superior predictive performance, achieving an accuracy of 98.66% along with consistently high precision and recall across nearly all classes. These results suggest that Naïve Bayes is more effective for predictive tasks involving sleep quality, while C4.5 remains highly valuable when the goal is to interpret variable interactions and decision rules. Overall, this research highlights the potential of data mining techniques in health informatics, particularly in improving the understanding and prediction of sleep quality, which in turn can contribute to better prevention and management of sleep-related health issues.

The application of interactive media in classifying Indonesian poetry based on stylistic devices aims to enhance students' understanding in identifying and analyzing the stylistic elements used in poetry. Stylistic devices such as metaphors, personification, hyperbole, and similes play a significant role in enriching the meaning of poetry and deepening the reader’s interpretation. The use of interactive media offers a more dynamic and effective approach compared to traditional methods. Through digital platforms and interactive learning applications, students can become more engaged in an active and enjoyable learning process, making it easier for them to recognize and comprehend various stylistic devices in poetry. This media also supports diverse learning styles, such as visual, auditory, and kinesthetic, enabling students to learn according to their preferences. This study aims to explore the effectiveness of interactive media in teaching the classification of stylistic devices in poetry and how it can enrich the learning experience while improving students' analytical skills in literary works. It is hoped that the application of interactive media can create a more engaging learning environment, boost students' motivation, and deepen their understanding of Indonesian poetry.

Typhoid fever is an infectious disease caused by the bacterium Salmonella typhi, commonly found in developing countries, including Indonesia. Prompt and accurate treatment is crucial to prevent serious complications in patients. One way to assist in diagnosing typhoid fever is by applying machine learning methods to classify patient data. The Naive Bayes method is one of the machine learning algorithms frequently used in medical data classification due to its strong ability to handle large and complex datasets. This article discusses the application of the Naive Bayes method for classifying typhoid patient data at Rantauprapat General Hospital (RSUD Rantauprapat). By utilizing medical data that includes clinical symptoms, laboratory test results, and patients’ medical histories, the Naive Bayes model can provide fairly accurate predictions regarding the likelihood of a person having typhoid fever. The research findings indicate that Naive Bayes is reliable in predicting typhoid diagnoses with adequate accuracy, thereby supporting healthcare professionals in making faster and more precise decisions. It is expected that the implementation of this method can accelerate the diagnostic process and improve the quality of healthcare services at RSUD Rantauprapat, as well as in other regions.

Child developmental delay is a public health issue that needs to be identified early to prevent long-term impacts on children’s quality of life. In Rantau Prapat Sub-district, cases are still found among toddlers with undernutrition, incomplete immunizations, and suboptimal developmental stimulation, which may pose risks of growth and developmental delays. This study aims to apply the Naive Bayes method in identifying child developmental delays based on health data collected through medical records and questionnaires. The research method includes data collection, pre-processing (cleaning, transformation, and normalization), classification using the Naive Bayes algorithm, and model validation with the k-fold cross-validation technique. The results showed that out of 150 toddler data samples, 30.7% experienced developmental delays, with the dominant influencing factors being nutritional status and immunization completeness. The Naive Bayes algorithm achieved an accuracy rate of 87.3% with a precision of 84.1%, recall of 85.7%, and F1-score of 84.9%. These findings demonstrate that Naive Bayes can be used as a decision support system in the early identification process of child developmental delays. Therefore, the results of this study are expected to assist healthcare workers, particularly midwives, in improving the quality of early detection and delivering more targeted interventions for children in the Rantau Prapat area.

The advances in the movie industry have led to the rapid rise of streaming platforms, becoming increasingly popular among people. With countries around the world constantly producing movies, subtitles play an essential role in making films more accessible to a broader audience. Subtitling strategies can be defined as a translation practice involving the display of written text, usually at the bottom of the screen, to convey the original dialogues of the speakers and the information contained in the soundtrack. This study uses data from the lk21 website and analyzes it using Gottlieb’s subtitling strategies theory, which consists of ten specific methods for audiovisual translation: Expansion, Paraphrase, Transfer, Imitation, Transcription, Deletion, Dislocation, Condensation, Decimation, and Resignation. A descriptive qualitative approach was adopted in this study with observation, classification, and note-taking of each subtitle, suitable for the research, and describing the subtitling strategies applied in the movie. The findings show that seven strategies were applied in the Indonesian subtitles, namely paraphrase, transfer, condensation, imitation, deletion, dislocation, and expansion. Among them, paraphrase emerged as the most frequently used strategy, likely because figurative or culture-specific expressions in the source text often required adaptation to preserve their meaning for Indonesian viewers. On the other hand, expansion was identified as the least applied strategy, suggesting that the original dialogues rarely demanded additional explanation. Overall, the strategic use of these methods contributed to producing subtitles that were coherent, culturally adapted, and highly accessible to the target audience.

The proliferation of fake news across digital platforms has raised critical concerns about information reliability. A notable example is the viral rumour falsely claiming that the Nigerian Minister of the Federal Capital Territory, Nyesom Wike, had collapsed at an event and was rushed to an undisclosed hospital an entirely fabricated claim that caused public confusion. While both traditional machine learning and deep learning approaches have been explored for automated fake news detection, many existing models have been limited to topic-specific datasets and often suffer from overfitting, especially on smaller datasets like ISOT. This study addresses these challenges by proposing a standalone Bidirectional Long Short-Term Memory (BiLSTM) model for fake news classification using the ISOT dataset. Unlike multi-modal frameworks such as the MM-FND model by state-of-the-art model, which achieved 96.3% accuracy, the proposed BiLSTM model achieved superior results with 98.98% accuracy, 98.22% precision, 99.65% recall, and a 98.93% F1-score. The model demonstrated balanced classification across both fake and real news and exhibited strong generalization capabilities. However, training and validation performance plots revealed signs of overfitting after epoch 2, suggesting the need for regularization in future work. This study contributes to the growing body of research on fake news detection by showcasing the efficacy of a focused, sequential deep learning model over more complex architectures, offering a practical, scalable, and robust solution to misinformation detection

Discipline is a crucial aspect of education that plays a significant role in shaping students’ character and sense of responsibility. However, the manual discipline assessment process at SDIP Baitul Maal presents several issues, such as inaccurate data, limited analysis capabilities, and difficulty for teachers and parents to monitor students effectively. With the advancement of information technology, digital systems offer a potential solution to improve the efficiency and objectivity of the evaluation process. This study aims to design and develop a web-based application for assessing student discipline by implementing the K-Means Clustering method optimized with the Elbow method. The system is designed to cluster students based on numerical data such as attendance, tardiness, neatness, and rule violations, allowing for more accurate classification of discipline levels. The results show that the system successfully groups students into clusters automatically and provides informative visualizations of the outcomes. Additionally, the system facilitates real-time monitoring and evaluation by school staff and parents through a user-friendly interface. Therefore, the application of the K-Means Elbow method proves effective in supporting decision-making within the educational environment. This research is expected to contribute to the digital transformation of school management and enhance the quality of student character development.

This study aims to analyze the implementation of Interpretation of Financial Accounting Standards (ISAK) 335 in the financial reporting of LP Ma’arif SMKS NU Pace Nganjuk as a non-profit entity. As an educational institution under a foundation, SMKS NU Pace is required to prepare financial statements that are transparent and accountable in accordance with applicable standards for non-profit entities. The research uses a descriptive qualitative approach with a case study method. Data were collected through interviews, observation, and documentation of existing financial reports. The findings indicate that the school has prepared reports such as BOS and BPOPP fund reports, as well as annual reports submitted to the foundation. However, the implementation of ISAK 335 remains basic and lacks detailed presentation. Key elements such as the classification of restricted and unrestricted net assets and disclosures in the Notes to Financial Statements (CALK) have not been fully understood or applied. The main obstacle is the limited technical understanding of ISAK 335 and the absence of specific accounting training for financial managers. It can be concluded that the implementation of ISAK 335 at LP Ma’arif SMKS NU Pace is still in its early stages and not yet optimal. Improving human resource capacity and establishing more standardized recording systems are essential.