Publication Search

This study presents the implementation of Transfer learning using the ResNet-18 architecture for classifying 10 musical instrument categories based on visual representations of audio signals. The audio waveform is transformed into image-like inputs appropriate for CNN processing, accompanied by data augmentation and ImageNet-standard normalization. ResNet-18 is utilized due to its efficient feature extraction capability enabled by residual blocks, which help overcome vanishing gradient issues. The model was trained for 10 Epochs using the AdamW optimizer and Cross-Entropy Loss. Experimental results show that the model achieved a maximum validation accuracy of 77.35%, with a stable downward trend in training loss, indicating effective feature learning. However, several misclassification cases were observed, particularly among instruments with similar spectral characteristics, such as drum–violin and tabla–sitar. These findings demonstrate that while ResNet-18 performs reliably for musical instrument classification, further improvements remain possible through deeper architectures like ResNet-50, more comprehensive hyperparameter optimization, and the use of richer audio representations such as Mel-Spectrograms. This research provides an essential foundation for developing automated music analysis systems powered by Deep Learning.

This research aims to analyze the role of social support on learning motivation among migrant students at the Faculty of Economics, Jakarta State University class of 2022. Migrant students often face challenges in the form of adapting to a new environment, which can influence their learning motivation. Social support, whether from family, friends, or the campus environment, is an important factor in helping this adaptation process. This research identifies types of social support such as emotional, instrumental, informational and self-esteem support, and their influence on learning motivation. Based on existing literature, social support can increase students' self-confidence, which has an impact on higher enthusiasm and motivation to learn. On the other hand, a lack of social support has the potential to cause stress, low self-confidence, and loss of motivation to learn. Using the case study method, this research examines the relationship between social support and the learning motivation of migrant students. The research results show that social support has a positive correlation with learning motivation. External factors, such as the campus environment and lecturers' teaching methods, as well as internal factors, such as personal interests and suitability of the field of study, also influence student learning motivation. This research is expected to provide new insights in increasing the learning motivation of migrant students through strengthening social support.

Landslide disasters can cause environmental damage, property loss, and cause deaths, disappearances, injuries, and displacement with various health problems in refugee camps such as infectious diseases and nutritional disorders. The level of disaster risk is determined by the student's potential and preparedness which can be known from the student's interpretation of landslide disaster management. Health education is a learning process that can change students' preparedness to be able to prepare action plans to reduce the impact of landslides. The aim of this research is to determine the effect of health education on landslide disaster management at SDN 1 Selo Boyolali. This type of research uses pre-experimental methods with a quantitative approach and a one group pretest-posttest research design. The sample consisted of 29 respondents using purposive sampling technique. Measuring student preparedness uses a preparedness questionnaire sheet with 25 questions using a Likert scale. The data analysis technique uses the Wilcoxon test with the research results showing a p-value of 0.000 (p<0.05), so that it can be concluded that there is an influence of landslide disaster management health education on student preparedness at SDN 1 Selo Boyolali.

Faculty The Faculty of HOLS Hospitality School faces challenges in manual administration, such as accumulating registration data, difficulty in data retrieval, risk of data loss, administrative inaccuracies, and time-consuming registration and attendance processes, increasing the risk of administrative errors. To address these issues, researchers designed a web-based information system using the waterfall model and PHP programming language. This system allows students to mark attendance using a QR code within a 5-meter radius, mentors to reschedule classes quickly and easily, and administrators to control administrative data accurately and efficiently. The result of this research is an information system product that facilitates academic administration management at HOLS Hospitality School, supports e-learning facilities, and simplifies coordination between students, mentors, and administrators.

The process of manually identifying fruits to determine ripe and unripe can affect the production and quality of the food and beverages themselves. The CNN method is able to group images and analyze images based on objects. Therefore, it is necessary to conduct research using the CNN method on the ripeness of strawberries. This study aims to determine the level of maturity of strawberries during harvest time. The accuracy graph shows that the model is not only capable of learning the training data well but can also generalize well to the validation data. In contrast, the validation accuracy graph starts from 0.825 in the 0th epoch and rises consistently until it reaches 0.975 in the 30th epoch. Both charts remained stable above those values throughout the training period. Overall, the development of the CNN model for the detection of strawberry ripeness resulted in excellent performance. The model achieved the lowest loss of 0.0383 and an accuracy as high as 98% on the validation data, demonstrating a strong ability to accurately predict between ripe and unripe strawberries.

This research discusses the implementation and evaluation of the Convolutional Neural Network (CNN) convolutional neural network model for classification of fruit types, specifically to differentiate between Banana and Papaya. The CNN model used consists of several convolutional, pooling, and fully connected (dense) layers designed to extract features and perform binary classification. Data augmentation is applied to the training set to increase data variation and prevent overfitting. The image data used is normalized to speed up training convergence. The model was trained using the Adam optimizer and the binary crossentropy loss function for 20 epochs. Performance evaluation was carried out using the validation set. The results show that the model is able to effectively classify fruit images with a high level of accuracy. Predictions are made by uploading images, resizing them, and normalizing them before using the model for predictions. The classification threshold was set at 0.4, where a predicted probability greater than or equal to 0.4 was classified as Banana and a probability less than 0.4 was classified as Papaya. This research shows that the CNN model can be used effectively for binary image classification tasks and can be extended to classify more types of fruit with appropriate data adjustments and model architecture.

Parental divorce significantly impacts early childhood development, especially in the digital age. Children may experience low self-esteem, difficulties in building relationships and socializing, aggressive behavior, learning difficulties, value confusion, deviant behavior, and loss of trust.Addressing these impacts requires digital parenting, open communication, moral education, and collaboration with educational institutions. Parents and stakeholders must understand the effects of divorce to provide appropriate support for children's healthy and happy growth. A qualitative case study at TK Negeri Bung Karno involved interviews with PAUD teachers and parents, as well as classroom observations. The findings showed that children tend to imitate negative parental behaviors. Through gradual guidance, children learned to distinguish between good and bad behavior, showing progress in emotional control, social interactions, and setting positive examples for peers.

Chronic diseases in humans are very difficult to detect visually, for example glaucoma, hypertension, diabetes, and others. So it takes a lot of time for further medical examination by visiting a health center or hospital. Therefore, this research aims to find a solution combining medical and computer science to classify quickly and precisely. Classifying eye images requires good features and characteristics so that disease images can be classified. This research uses the Deep Learning method, namely Convolutional Neural Network with MobileNet-V3 architecture which can extract features from large resolution images very well. This research resulted in accurate classification of images of chronic diseases Normal, Diabetes, Glucoma, Cataract, Age related macular degeneration, Hypertension, Pathalogical Myopia. uses the MobileNet-V3 architecture, with transfer learning reaching 81%, and loss only 0.4913.

This research aims to develop and apply a Convolutional Neural Network (CNN) algorithm to detect handwritten Roman numerals. Handwriting recognition is a classic challenge in the fields of image processing and machine learning, especially for less common characters such as Roman numerals. In this research, we use data augmentation techniques to increase the diversity and number of datasets used in model training, which is expected to increase model accuracy and generalization. The dataset used consists of 1,120 images for testing and 280 images for validation, each of which is divided into 14 classes of Roman numerals I, II, III, IV, V, VI, VII, VIII, IX, X, L, C, D , and M. Image data was created directly using simple software, namely Paint version 6.3. This research uses the Python programming language and Google Colab as a computing platform. Model training was carried out for 300 epochs and showed significant accuracy in the 150th to 300th iteration. The results at the 300th epoch show an accuracy of 0.9607 and a loss of 0.1162. The implementation of this algorithm shows significant potential in practical applications, such as in the fields of education and historical documentation. The conclusion of this research is that data augmentation is an effective technique to improve the performance of CNN models in detecting handwritten Roman numerals.

Bullying refers to deviant behavior that is often carried out consciously or unconsciously. Bullying is repeated aggressive and negative behavior towards someone with the aim of hurting him. Bullying is defined as negative actions carried out by a stronger party against a weaker party through the use or non-use of tools with the aim of creating physical and emotional pressure on the weaker party. The increase in bullying cases in schools shows how worrying the current condition of education in Indonesia is. The aim of this research is to find out whether bullying behavior has a significant influence on students' learning motivation and enthusiasm for learning. We hope this can help schools know how big the impact of bullying is at school and help victims overcome the problems and risks of bullying. The method used is the Systematic Literature Review or SRL method, namely selecting each journal or previous research and getting results in the form of references for models/ideas that will be developed in the future. The results of this research show that bullying has a big influence on students' learning motivation, this is characterized by not having the courage to participate actively in the learning process and showing low interest in lessons. Bullying behavior significantly affects elementary school students' learning motivation. Students who experience bullying tend to show lower levels of motivation, loss of interest in learning, and lack of confidence in their academic abilities.

The challenges of millennials cover various aspects such as economy, employment, property ownership and leadership. According to Glints, the first challenge millennials face is economic, caused by the impact of previous economic crises and capital losses. In the world of work, millennials are often associated with stigmatization in the workplace and cross-generational differences in working styles, which can be challenging. Property ownership is also a challenge due to the high consumption of millennials. Managers need to be able to adapt to change and not feel stressed as millennials are more likely to seek real work and get bored quickly in the workplace. Millennials also value their own experiences and strengths, so they need to focus on that. This research uses a qualitative approach. This article uses library inquiry techniques, or library investigation, to copy data and understand theories found in relevant literature. The data used for this purpose is undoubtedly a precedent study of related arguments, such as libraries, journals, and scientific articles. Christian Religious Education (CHE) plays a crucial role in holistic education. Through PAK, millennials are guided to cultivate Christian values such as love, peace, honesty and justice. It also plays a role in shaping a moral and virtuous character based on the example of Jesus Christ. In addition, PAK equips millennials with a strong moral and ethical foundation, helps them make responsible and moral decisions, and guides them to live their lives in accordance with Christian values. In today's digital era, Christian Religious Education (CHE) needs to adapt to the technology used in students' lives. This is important to ensure that learning is not boring and can engage students. In addition, Christian Religious Education should also shape students' self-image according to the values of biblical truth. In this way, students can develop a strong character based on Christian faith. PAK in its ministry should encourage every believer, from children to adults, so that they can use and optimize their intelligence/knowledge for the common good Christian Religious Education (CHE) in the modern era faces the challenge of educating the millennial generation which has unique characteristics. This generation is accustomed to technology and has high self-confidence. Therefore, PAK needs to adapt to the technology used in students' lives to ensure that learning is not boring and engaging. In addition, PAK should focus on building students' self-awareness according to the values of Biblical truth and Christian Religious education teaching. This is important to help students develop a strong character based on the Christian faith. In this way, PAK can have a positive impact on the millennial generation and help them become people of faith and morals in an increasingly modern world.    

In developing an independent curriculum in Indonesia, one of them cannot be separated from the role of philosophy of science, which in curriculum development is the spirit or foundation of education which must be evaluated, studied periodically, innovative and dynamic following changing times, such as the philosophical Ki Hajar Dewantara who stated that education is appropriate. the nature of nature and the nature of the times, meaning that education develops according to the times. Of course, all the changes and developments that occur in the educational curriculum cannot be separated from the results of several studies on how the implementation of this independent curriculum can overcome learning loss, the effects of the Covid pandemic plus the global challenges of change from era 4.0 to Society 5.0, which has the impact that today's generation needs. have the ability to think critically, creatively and innovatively. In the independent curriculum, it can be seen that the superiority of its characteristics is that the material is simpler and more in-depth and there is a significant reduction in material. The materials presented are limited to essential material. Reducing material provides an opportunity for students to explore the material more comprehensively. Next is more independence, which means that in an independent curriculum, students are given more freedom to choose the subjects they are interested in according to their interests, talents, potential and aspirations. Meanwhile, teachers can teach according to the stages of student achievement and development. So it is more relevant and interactive. Students are guided to be able to collaborate, work together to form groups to identify problems that are becoming issues to strengthen the profile of Pancasila students, namely faith and devotion to God Almighty and noble character, global diversity, mutual cooperation, independence, critical reasoning and creativity. The essence of this independent curriculum is independent learning. This aims to enable students to explore their respective interests, potential and talents.

One of the government programs created to realize MBKM (Free Learning and Independent Campus) activities is the Teaching Campus. This program is intended for students as agents of change to increase human resources by contributing and providing innovation to educational problems, especially in learning activities. This campus teaching activity includes several activities regarding literacy and numeracy, adapting technology to more innovative methods. The Covid-19 pandemic has caused changes in students' learning patterns (Learning Loss), which has had a major impact on children's interest in learning, understanding of the material and learning outcomes. In connection with existing problems, the aim of writing this journal is to have a direct impact on holding campus teaching programs. Research or studies from this journal were carried out using a qualitative approach in the form of collecting data on learning outcomes during the program. The results of research and studies show that there is an increase in students' interest in learning by using learning media that is more exciting and fun. Apart from that, this learning method is very effective in increasing the value of the learning outcomes of class V students at SDN Simokerto I / 134. Good learning outcomes come from good learning methods, because interest in learning is proportional to the understanding gained too.

Potential weaknesses faced by the company. So it is necessary to consider the aspects of process inefficiency, risk of work errors, delays in information, lack of transparency, high operational costs, vulnerability to data loss and difficulties in monitoring performance. In the current era of globalization and digitization, companies face pressure to increase operational efficiency and accountability in presenting financial information. By using the PALS (Pasticipatory Action Learning System) method, where this method basically involves accounting staff in the process of upgrading the Enterprise Service Bus (ESB) system from 30 accounting staff and 5 staff will be taken for the interview sample in this research. From the problems currently faced by companies in managing financial reports, they have not been able to carry out systematic recording. With better data integration using ESB, a company's various systems and applications can be connected and communicate well. In real-time reporting compared to manual (conventional) recording which requires processing time, ESB allows for better real-time financial reports.

Quality education requires a quality learning process. Quality learning is certainly designed in learning activities that use learning methods that are in accordance with the eyes taught by the teacher so as to assist the teacher in giving directions so that students can understand the content delivered by the teacher. In reality, education in Indonesia has decreased the quality of learning in the context of an independent curriculum. There are research findings called learning loss and learning gap devices in this problem. So it is expected that subject teachers must be able to restore the declining level of learning quality.  So researchers use a problem-based learning model. Where this problem-based learning model can solve problems and increase students' knowledge. The purpose of the study was to analyze the implementation of problem-based learning in improving the quality of PPKn learning at SMP Negeri 19 Malang. The method used in this research is qualitative case study method. Data collection techniques based on observation, interviews and document research. The result of this study is the application of the ICT-assisted problem-based learning model showing that a significant change because students are directed to examine problems in learning activities using problem-based teaching methods that have a positive impact on student achievement outcomes. Based on the results of the research, researchers can conclude that the application of ICT-assisted problem-based learning models in class VIII PPKn subjects of SMP Negeri 19 Malang students can solve problems in groups given by teachers.

Plant disease is one of the problems in the world of agriculture. Early identification of plant diseases can reduce the risk of loss, so automation is needed to speed up identification. This study proposes a custom-designed convolutional neural network (CNN) model for plant disease recognition. The proposed CNN model is not complex and lightweight, so it can be implemented in model applications. The proposed CNN model consists of 12 CNN layers, which consist of eight layers for feature extraction and four layers as classifiers. Based on the experimental results of a plant disease dataset consisting of 38 classes with a total of 87,867 image records. The proposed model can get high performance and not overfitting, with 97%, 98%, 97% and 97%, respectively, for accuracy, precision, recall and f1-score. The performance of the proposed model is also better than some popular pre-trained models, such as InceptionV3 and MobileNetV2. The proposed model can also work well when implemented in mobile applications.

Convolutional neural network (CNN) is a deep learning (DL) model that has significantly contributed to medical systems because it is very useful in digital image processing. However, CNN has several limitations, such as being prone to overfitting, not being properly trained if there is data duplication, and can cause unwanted results if there is an imbalance in the amount of data in each class. Data augmentation techniques are used to overcome overfitting, eliminate data duplication, and random under sampling methods to balance the amount of data in each class, to overcome these problems. In addition, if the CNN model is not designed properly, the computation is less efficient. Research has proved that data augmentation can prevent or overcome overfitting, eliminating duplicate data can make the model more stable, and balancing the amount of data makes the model unbiased and easy to learn new data as evidenced through model evaluation and testing. The results also show that the custom convolutional neural network model is the best model compared to ResNet50 and VGG19 in terms of accuracy, precision, recall, F1-score, loss performance, and computation time efficiency

Character education has been rolled out since 2010 and is still valid today. In character education there are several values, one of which is responsibility. The character of responsibility is very important so that the school becomes a self-development institution to achieve success. Depletion of awareness and even loss of the character of responsibility in students which hinders learning activities so that this should not happen (Kamila, 2013). This study focuses on definitions, benefits, and strategies in learning to improve the character of student responsibility in schools. The results of this study include: 1) character refers to a series of attitudes, behaviors, motivations, and skills, while the character of responsibility is the attitude and behavior of a person to carry out the duties and obligations that he should do to yourself, society, environment (natural, social, and cultural), the State and God Almighty; 2) the benefits of the character of responsibility are that it can improve the quality of learning in the classroom and can improve school quality standards; 3) the character of responsibility can be done with a mind mapping strategy, inquiring mind what to know, CRH, NHT, and Quiz Team. This study concludes that the character of responsibility and learning must be carried out as one of the character education in schools through several strategies.

Dampak yang paling menonjol pada pandemi Covid-19 adalah penutupan sekolah dan Pembelajaran Jarak Jauh (PJJ) daring. Guru memberikan bahan pembelajaran dengan memanfaatkan jaringan internet. Guru hanya memberikan bahan pembelajaran melalui zoom padahal peran guru di sekolah bukan hanya memberikan bahan pembelajaran tetapi juga mengasah bakat minat anak. Setiap anak itu unik dengan kemampuan yang luar biasa untuk dikembangkan sesuai dengan potensinya. Guru dalam mengembangkan bahan pembelajaran juga harus memperhatikan potensi anak.Tujuan dari jurnal non penelitian ini adalah peran guru untuk mengembangkan bakat minat siswa di sekolah. Memaksimalkan potensi anak berkaitan dengan bakat dan minat anak juga perlu berkolaborasi dengan orang tua untuk mewujudkan bakat tersebut.

The most noticeable impacts of the Covid-19 pandemic are school closures and online distance learning (PJJ). Online learning uses the Internet to provide teaching materials to students. Barriers to online learning are felt by both teachers and students, with streaming learning via Zoom sometimes having unstable internet connections. Another obstacle to online learning between teachers and students is the teacher-student relationship. You look nervous because you get stuck in the Zoom application. A barrier to online learning between students and parents is that it is difficult for working parents to tutor their children because they are working. The impact of online learning itself takes the form of reduced learning ability and changes in student behavior. The purpose of this non-research journal is that schools need new strategies to meet their post-pandemic educational goals. In order to overcome this, it is necessary to develop human resources by implementing education that reflects the cultural characteristics of the country. Education must respond not only to global requirements, but also to developments in digital technology.