Publication Search

The rapid advancement of deep learning technologies has significantly transformed healthcare analytics, particularly in medical data prediction and classification. This study proposes a hybrid Convolutional Neural Network–Long Short-Term Memory (CNN–LSTM) framework for multi-modal healthcare data analysis, integrating medical imaging, structured electronic health records (EHRs), and IoT-generated time-series physiological signals. The proposed architecture combines spatial feature extraction through CNN with temporal dependency modeling via LSTM to enhance predictive accuracy and clinical decision support. A quantitative experimental design was employed, utilizing multi-source healthcare datasets that underwent preprocessing, normalization, and feature engineering prior to model training. The performance of the hybrid model was evaluated using Accuracy, Precision, Recall, F1-Score, AUC-ROC, and Mean Absolute Error (MAE), and compared with conventional machine learning models and standalone deep learning architectures. Experimental results demonstrate that the proposed CNN–LSTM model achieves superior performance, with improved classification accuracy and reduced prediction error, while maintaining strong generalization capability. The findings indicate that integrating spatial and temporal feature learning significantly enhances disease detection, risk stratification, and personalized treatment planning. This approach supports the development of intelligent clinical decision support systems and scalable smart healthcare environments. The proposed framework offers a reliable and efficient solution for advanced healthcare analytics in IoT-enabled systems.

This study explores the application of green hydrogen technology for industrial decarbonization, focusing on its techno-economic and environmental feasibility. A quantitative approach was used, incorporating system modeling of a solar-based hydrogen production system combined with electrolyzers. The techno-economic assessment involved calculating the Levelized Cost of Hydrogen (LCOH), estimating capital and operational expenditures (CAPEX and OPEX), and evaluating the system's energy efficiency and hydrogen output. The environmental impact was analyzed using Life Cycle Assessment (LCA), comparing the carbon footprint of green hydrogen with fossil-based hydrogen. The results reveal that green hydrogen can reduce carbon emissions by up to 60% compared to fossil hydrogen, primarily due to the use of renewable energy for production. Additionally, the study found significant improvements in energy efficiency as electrolyzer performance and solar capacity increased. The LCOH is expected to decrease steadily as solar panel and electrolyzer prices continue to fall, enhancing the competitiveness of green hydrogen in the energy market. The findings also highlight the potential for heavy industries, such as cement and steel production, to transition from fossil fuels to green hydrogen, contributing to a cleaner industrial energy mix. This transition presents both environmental and economic benefits, with long-term savings from reduced fossil fuel dependency and lower production costs.

A study was conducted to identify the relationship between height and weight. A researcher claimed that on average, when height increases, weight increases. This is called the Benn parameter. Weight can be adjusted for height using the Benn parameter (kg/mB), where B is the power that minimizes the correlation with height. To investigate how the Benn parameter changes across age and differs between genders. Using linear regression. Multilevel modeling was used to develop a history showing how Benn's parameters change with age from childhood through mid-adulthood and into older age. The data used information from a sample of 40 adults and 40 children under five. Height was measured in centimeters, weight was measured in kilograms. At alpha equal to 0.05 (α=0.05), the hypothesis is that there is a relationship between height and weight. This is the data, which is height and weight. And here is the hypothesis.  (Hnull) there is no relationship between height and weight, ρ equals 0. And ,, there is a relationship between height and weight, ρ does not equal 0. And we test to claim that .

The improvement of performance for Micro and Small Enterprises (MSEs) is a must to achieve business performance growth. To achieve business performance successfully, the MSEs partly need to investigate factors that influence it. This study aims to examine whether or not financial access, digitalization adoption, product innovation, and market orientation influence MSEs' business performance. The source of data was obtained by distributing questionnaires to 135 women who engaged in MSEs in Belitung Regency as one of the main locations of women MSEs in Indonesia. These data were then analyzed statistically by applying the Partial Least Square-Structural Equation Modeling (PLS-SEM. The main findings of this research are: (i) the path coefficient of the variables financial access, product innovation, and market orientation has a positive and significant effect, (ii) the path coefficient of digitalization adoption has a positive but not significant effect on MSE business performance. The results of this research provide input for women's MSEs in Belitung Regency to be able to improve their market orientation to overcome marketing obstacles that may be experienced and to improve the business performance of women's MSEs in Belitung Regency.

This research aims to analyze the effect of using Web Wordwall in training students' computational thinking skills in mathematics learning with a qualitative approach. The research subject is the Director General of Teachers and Education Personnel, while the object is education that is liberating and pro-student. Data obtained from online and offline information sources is recorded in a structured manner, then analyzed to be synthesized into a systematic conceptual framework. Conclusions are drawn by taking the essence of the conceptual framework that can answer the research questions posed. The research method used is a case study with a focus on participatory observation, in-depth interviews and document analysis. Data was collected through literature study and analysis of learning materials. Data analysis was carried out using an inductive approach, by looking for patterns of findings and main concepts from the data collected.

This research aims to analyze the effect of using Web Wordwall in training students' computational thinking skills in mathematics learning with a qualitative approach. The research subject is the Director General of Teachers and Education Personnel, while the object is education that is liberating and pro-student. Data obtained from online and offline information sources is recorded in a structured manner, then analyzed to be synthesized into a systematic conceptual framework. Conclusions are drawn by taking the essence of the conceptual framework that can answer the research questions posed. The research method used is a case study with a focus on participatory observation, in-depth interviews and document analysis. Data was collected through literature study and analysis of learning materials. Data analysis was carried out using an inductive approach, by looking for patterns of findings and main concepts from the data collected.

This research examines employees of Perum Jasa Tirta I Semarang. This research aims to determine the influence of cyberloafing and leader member exchange on innovative work behavior through employee performance. The research method in this research is a quantitative research method. The population in this study was all employees with a total of 10 employees who worked at Perum Jasa Tirta I Semarang, while the sample in this study was 180 respondents. Determining the sample size used a non-probability sampling technique because the population was relatively small, so the technique taken was saturated sampling (census). Data analysis in this research used the SmartPLS version 3 program. The data analysis technique used was the Structural Equation Modeling (SEM) equation model. The results of the analysis and discussion show that the perceived quality variable influences Cyberloafing, influences Innovative Work Behavior, Leader Member Exchange influences Innovative Work Behavior, Cyberloafing influences Employee Performance, Leader Member Exchange influences Employee Performance, Employee Performance influences Innovative Work Behavior, Cyberloafing influences on Innovative Work Behavior is partially mediated by Employee Performance, and Leader Member Exchange influences Innovative Work Behavior of Perum Jasa Tirta I Semarang employees which is partially mediated by Employee Performance.

This research was motivated by the phenomenon of guest repurchase intention allegedly due to the quality of service, brand preferences, and prices at Warung Nasi Kelok Desa Talogondan. This study aims to analyze the role of satisfaction in mediating the effect of service quality, brand preference, and price on repurchase intention in Warung Nasi Kelok Talogon Village and Padang Pariaman Regency. This type of research is quantitative research with a causal associative approach. The population in this study is rice stall consumers who have visited Warung Nasi Kelok Talogon Village and Padang Pariaman Regency with a sample of 222 respondents. The research instrument used in this study was a questionnaire (questionnaire) arranged according to the Likert scale. Instrument trials use validity tests, reliability tests. Data analysis technique using Structural Equation Modeling with Partial Least Square (SEM-PLS). The result of the study is that there is a positive and significant influence between service quality variables and consumer satisfaction variables. There is a positive and significant influence between brand preference variables and consumer satisfaction variables. There is a positive and significant influence between price variables and consumer satisfaction variables. There is a positive and significant influence between the consumer satisfaction variable and the repurchase intention variable. There is an influence of customer satisfaction as a positive and significant mediating variable between service quality and repurchase intention. There is an influence of consumer satisfaction as a positive and significant mediating variable between brand preference for repurchase intention and path coefficient. There is an influence of consumer satisfaction as a positive and significant mediating variable between price and repurchase intention and path coefficient.

Objective – This research clarifies the factors that encourage creative economic inclusiveness for MSME actors by examining the effect of optimizing MSME empowerment, the role of digital technology literacy and ease of access for MSME actors. This research also examines analysis direct influence optimization empowerment on digital technology literacy and ease of access, the influence of digital technology literacy on ease of access and the mediating role of digital technology literacy and ease of access. Design/methodology/pe approach – Data was taken from an online questionnaire of respondents, namely MSME actors in Purwakarta Regency. Partial least squares structural equation modeling and mediation analysis using the bootstrap method. Findings – Optimizing empowerment and ease of access has a significant direct impact on the inclusiveness of the creative economy. Both optimizing empowerment and digital technology literacy have a direct, positive and significant influence on ease of access. Optimizing empowerment has a positive effect on digital technology literacy. Mediation analysis revealed that ease of access was unable to mediate the influence between optimizing empowerment and inclusiveness of the creative economy and between digital technology literacy and inclusiveness of the creative economy. Research limitations/implications – This research focuses on three factors that encourage creative economic inclusiveness of MSME actors in Purwakarta Regency. Originality/value – This study offers new insights into the factors that encourage the inclusiveness of the creative economy of MSME players. These findings reveal the importance of optimizing MSME empowerment and encouraging digital technology literacy for creative economic inclusivity among MSME players.

The development of technology is something that cannot be separated from humans at this time, it is based on technology that is increasingly growing. But on the problems faced in the search for facilities and services private tutoring is still an information manual so that it is difficult for the community to find facilities and services for private tutoring especially in non-academic fields. Then it needs to be specifically designed as a search system for facilities and private tutoring services through web-based media. Thus the community is not difficult to find a place or non-academic field private tutoring services because in this system there is a latitude of maps that will show the location and there are display fees and online payments via a bank account. In designing the system the author observes the company and the community as well as a bibliography that aims to be a benchmark. The author uses the Business Model Canvas (BMC) method with the support of system requirements elicitation, as well as modeling the Unified Modeling Language (UML) system to visually implement the Hypertext Preprocessor (PHP) program language with the MySQL database as a database server. With the design of a search system for facilities and private tutoring services, it can make it easier for people to find effective and efficient facilities or services in non-academic fields.

Energy efficiency is a critical concern in wireless sensor networks (WSNs) due to the limited power resources available in sensor nodes. Prolonging network lifespan while ensuring reliable data transmission is essential for successful deployment in various applications, such as environmental monitoring, military operations, and industrial automation. This paper presents a mathematical model designed to optimize energy consumption across various nodes in WSNs. By implementing simulations and analyzing data from these models, the study demonstrates significant improvements in extending network lifespan while maintaining reliable data throughput. The findings contribute valuable insights into energy management for large-scale sensor deployments.

Power systems are critical infrastructure that face significant challenges due to increasing demand and inherent complexity. Predicting failures in power systems is crucial for enhancing grid reliability, minimizing downtime, and optimizing maintenance processes. This study evaluates various deep learning models, specifically convolutional neural networks (CNN), recurrent neural networks (RNN), and transformer models, for predicting power system failures. By analyzing these models’ performance metrics on historical power grid data, the study provides insights into the strengths and weaknesses of each approach. The findings contribute to the development of more robust predictive models for power system reliability.

Software Modeling is a scientific discipline for studying forms of software modeling that are used as part of the software development stages in a structured and object-oriented manner. This software engineering modeling is presented in Requirements Analysis and described using the Unified Modeling Language (UML) method. UML (Unified Modeling Language) is a visual modeling method used as a means of designing object-oriented systems. UML diagrams that are often used are Use Case Diagrams, Activity Diagrams, Sequence Diagrams, Class Diagrams, Statemachine Diagrams and Component Diagrams. There are several applications that support UML creation, one of which is Draw.io. Draw.io is a website specifically designed for depicting UML diagrams online. It has a very responsive display and is integrated with Google's file storage service, namely Google Drive, so draw.io is an alternative for creating UML diagrams in a shorter time. On this basis, in the Community Service (PkM) program activities on this occasion, UML creation activities were carried out using the draw.io application and assistance in using UML in analyzing needs is one of the right solutions for solving problems in order to expand knowledge, simplify and improve the quality of education. and will overcome problems, time and socio-economic

This research examines the employees of UKM Batik Lasem Sekar Mulyo, Rembang. This research aims to determine the influence of the role of core self-evaluation and empowering leadership on employee creativity through knowledge sharing. The research method in this research is a quantitative research method. The population in this study were all employees with a total of 95 employees who worked at UKM Batik Lasem Sekar Mulyo, Rembang, while the sample in this study was 95 respondents. Determining the sample size used a non-probability sampling technique because the population was relatively small, so the technique taken was saturated sampling (census). Data analysis in this research used the SmartPLS version 3 program. The data analysis technique used was the Structural Equation Modeling (SEM) equation model. The results of the analysis and discussion show that the quality perception variable influences Core Self-Evaluation and influences Employee Creativity, Empowering Leadership influences Employee Creativity, Core Self-Evaluation influences Knowledge Sharing, Empowering Leadership influences Knowledge Sharing, Knowledge sharing influences Employee Creativity, Core Self-Evaluation has an effect on Employee Creativity, partially mediated by Knowledge sharing, and Empowering Leadership has an effect on Employee Creativity in Employees of UKM Batik Lasem Sekar Mulyo, Rembang, which is partially mediated by Knowledge sharing.

This study explores the application of stochastic differential equations (SDEs) in modeling population dynamics. By introducing randomness into differential equations, we capture the inherent uncertainties in environmental and genetic factors affecting population growth. We derive analytical solutions for specific cases and provide numerical simulations to demonstrate how SDEs enhance predictions in ecological modeling. Our findings suggest that stochastic models provide a more robust framework for understanding population fluctuations in uncertain environments.

This paper compares various machine learning models in their ability to predict financial trends, with a focus on time-series analysis. We evaluate models such as linear regression, decision trees, support vector machines, and deep learning, measuring their performance based on accuracy, computational cost, and interpretability. Our results reveal that deep learning models offer superior accuracy but are less interpretable, while simpler models, though less accurate, provide better insight into the underlying data. This research provides guidelines for selecting suitable models based on specific financial applications.

This paper introduces a hybrid optimization approach that combines genetic algorithms with gradient descent for effective nonlinear function approximation in highdimensional data. Traditional methods struggle with computational efficiency and accuracy in such complex spaces. By integrating genetic algorithms to provide a global search strategy with gradient descent for finetuning, the proposed method achieves faster convergence and improved accuracy. Simulations and case studies demonstrate its effectiveness in applications like data mining, image recognition, and financial modeling.

Web novels have gained popularity in recent years as a form of literature that is published and consumed online via specialized platforms. User generated comments and reviews play an important role in the web novel platform, providing valuable insight into reader sentiment and feedback. Manually analyzing sentiment from a large number ofcomments would be very time consuming, an efficient automated approach was required. This study uses the Latent Dirichlet Allocation (LDA) method to identify sentiment patterns (positive, negative, neutral) in user comments on web novels and analyze their distribution as a whole. LDA, originally designed for topic modeling, has proven effective in sentiment analysis, helping to group comments into relevant topics and uncover general sentiments related to each topic. This study aims to use the LDA method to identify sentiment patterns (positive, negative, or neutral) in user comments on web novels and analyze the distribution of sentiment as a whole. The results show the effectiveness of LDA in sentiment analysis, achieving quite good results, with 72% accuracy, 80% precision, 72% recall, and 65% F1 score.

This research aims to determine and try to analyze the influence of discipline and cooperation on work productivity with job satisfaction as an intervening variable at PT. Wilmar Nabati Padang. This research is a Structural Equation Modeling (SEM) model with the SmartPLS 3.0 analysis tool. The population and sample of this research are employees of PT. Wilmar Nabati Padang. The research results found that there is an influence between discipline and job satisfaction. There is an influence between Collaboration on Job Satisfaction. There is an influence between discipline and work productivity. There is no influence between Collaboration on Work Productivity. There is an influence between Job Satisfaction and Work Productivity. Job Satisfaction mediates Discipline on Work Productivity. Job Satisfaction mediates Collaboration on Work Productivity.

This study was conducted to evaluate the effectiveness of brand image in mediating the impact of word of mouth (WOM) on the desire to purchase Mie Sedaap Korean Spicy Chicken products in the Tangerang City area. Through a survey involving 200 individuals selected based on purposive sampling, this research utilized the Partial Least Squares Structural Equation Modeling (PLS-SEM) model to process and analyze the data obtained. The results found indicate a positive impact of WOM on brand image and purchase intention, with brand image also functioning as a significant mediator in linking WOM with purchase intention. The conclusion of these findings underscores the need for effective marketing strategies in utilizing WOM to strengthen brand image, which in turn has the potential to increase purchase intentions among consumers.