Nuke Puji Lestari Santoso
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Nuke Puji Lestari Santoso has a total of 80 co-authors. Here are the top co-authors by number of shared publications:
- Adam Faturahman
Universitas Raharja
3 shared publications - Qurotul Aini
Universitas Raharja
3 shared publications - Untung Rahardja
Universiti Teknologi Malaysia
3 shared publications Aman Jaiswal
Karachi Institute of Economics and Technology
2 shared publications- Anandha Fitriani
University of Raharja
2 shared publications - Bhupesh Rawat
Graphic Era Hill University
2 shared publications
Latest Publications
- Open Access
- 30 October 2026
- Published by iLearning Journal Center in Journal of Computer Science and Technology Application
- Vol. 3  (2) , 171-184
Deep Learning Enabled Security Monitoring for Intrusion Detection in Smart Campus Networks
The increasing complexity of smart campus networks has heightened the need for advanced cybersecurity measures to protect sensitive data and ensure seamless operations. Traditional Intrusion Detection Systems (IDS) often struggle to cope with the dynamic and heterogeneous nature of network traffic in smart campus environments, necessitating the development of more effective solutions. This study aims to propose a deep learning-based intrusion detection system for smart campus networks, utilizing a Hybrid CNN-LSTM model to enhance security monitoring. The proposed methodology integrates Convolutional Neural Networks (CNN) for feature extraction and Long Short-Term Memory (LSTM) networks for capturing temporal dependencies in network traffic. The model was trained and evaluated on a combination of publicly available datasets and simulated smart campus data, measuring performance through key metrics such as accuracy, precision, recall, and F1-score. Results show that the Hybrid CNN-LSTM model outperforms traditional machine learning models, achieving an accuracy of 97.3% and a ROC-AUC of 0.99, demonstrating superior detection of both known and unknown intrusions. The findings suggest that deep learning models, especially when tailored to smart campus contexts, offer significant advantages in real-time threat detection and adaptive learning. This research contributes to the growing body of knowledge on AI-driven network security and provides practical insights for improving cybersecurity infrastructures in higher education institutions.
- 30 October 2026
- Published by iLearning Journal Center in Journal of Computer Science and Technology Application
- Vol. 3  (2) , 171-184
Open AccessDeep Learning Enabled Security Monitoring for Intrusion Detection in Smart Campus Networks
The increasing complexity of smart campus networks has heightened the need for advanced cybersecurity measures to protect sensitive data and ensure seamless operations. Traditional Intrusion Detection Systems (IDS) often struggle to cope with the dynamic and heterogeneous nature of network traffic in smart campus environments, necessitating the development of more effective solutions. This study aims to propose a deep learning-based intrusion detection system for smart campus networks, utilizing a Hybrid CNN-LSTM model to enhance security monitoring. The proposed methodology integrates Convolutional Neural Networks (CNN) for feature extraction and Long Short-Term Memory (LSTM) networks for capturing temporal dependencies in network traffic. The model was trained and evaluated on a combination of publicly available datasets and simulated smart campus data, measuring performance through key metrics such as accuracy, precision, recall, and F1-score. Results show that the Hybrid CNN-LSTM model outperforms traditional machine learning models, achieving an accuracy of 97.3% and a ROC-AUC of 0.99, demonstrating superior detection of both known and unknown intrusions. The findings suggest that deep learning models, especially when tailored to smart campus contexts, offer significant advantages in real-time threat detection and adaptive learning. This research contributes to the growing body of knowledge on AI-driven network security and provides practical insights for improving cybersecurity infrastructures in higher education institutions.
- Open Access
- 28 October 2026
- Published by iLearning Journal Center in International Transactions on Artificial Intelligence (ITALIC)
- Vol. 5  (1) , 35-46
Machine Learning-Based Prediction of Electric Vehicle Charging Demand for Sustainable Transportation
The increasing adoption of Electric Vehicles (EV) presents challenges in managing uncertain charging demand, optimizing infrastructure utilization, and maintaining energy system stability. This study aims to develop and evaluate a machine learning-based prediction framework for estimating EV charging energy demand and identifying the most effective predictive algorithm to support sustainable transportation planning. The proposed framework employs historical EV charging observations and contextual variables, including charging duration, charging time, day of the week, ambient temperature, charging frequency, and historical energy consumption. Three supervised machine learning algorithms, namely Random Forest (RF), Support Vector Regression (SVR), and Extreme Gradient Boosting (XGBoost), are comparatively evaluated. Predictive performance is assessed using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination (R²). The reported experimental findings indicate that XGBoost achieves the highest predictive accuracy, with an RMSE of 2.84 kWh, MAE of 1.97 kWh, MAPE of 6.21%, and R² of 0.941. However, the comparative performance metrics for RF and SVR must be reported in the main Results section to substantiate the claimed superiority of XGBoost. The proposed framework demonstrates the potential of machine learning to identify charging demand patterns using historical charging, temporal, and environmental information. Subject to further experimental verification, the findings may support data-driven charging infrastructure management, electricity resource allocation, and sustainable transportation planning.
- 28 October 2026
- Published by iLearning Journal Center in International Transactions on Artificial Intelligence (ITALIC)
- Vol. 5  (1) , 35-46
Open AccessMachine Learning-Based Prediction of Electric Vehicle Charging Demand for Sustainable Transportation
The increasing adoption of Electric Vehicles (EV) presents challenges in managing uncertain charging demand, optimizing infrastructure utilization, and maintaining energy system stability. This study aims to develop and evaluate a machine learning-based prediction framework for estimating EV charging energy demand and identifying the most effective predictive algorithm to support sustainable transportation planning. The proposed framework employs historical EV charging observations and contextual variables, including charging duration, charging time, day of the week, ambient temperature, charging frequency, and historical energy consumption. Three supervised machine learning algorithms, namely Random Forest (RF), Support Vector Regression (SVR), and Extreme Gradient Boosting (XGBoost), are comparatively evaluated. Predictive performance is assessed using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination (R²). The reported experimental findings indicate that XGBoost achieves the highest predictive accuracy, with an RMSE of 2.84 kWh, MAE of 1.97 kWh, MAPE of 6.21%, and R² of 0.941. However, the comparative performance metrics for RF and SVR must be reported in the main Results section to substantiate the claimed superiority of XGBoost. The proposed framework demonstrates the potential of machine learning to identify charging demand patterns using historical charging, temporal, and environmental information. Subject to further experimental verification, the findings may support data-driven charging infrastructure management, electricity resource allocation, and sustainable transportation planning.
- Open Access
- 31 August 2026
- Published by Sinar Mentari Sundara in Smart Human-centered Emerging Research in Machine Intelligence
- Vol. 2  (2) , 110-120
Architectural Design of a Deep Learning Framework for Real-Time Urban Air Quality Anomaly Detection
The background of this research addresses the escalating environmental crisis caused by rapid urbanization and the urgent need for advanced technological interventions to monitor atmospheric conditions. Rapid industrial growth has led to severe air pollution which directly threatens public health and sustainable development across the globe. Therefore the primary object of this study is the architectural design of a highly efficient predictive framework capable of identifying sudden spikes in air pollution levels across smart city infrastructures. To achieve this the proposed method utilizes a novel deep learning architecture that synergizes one dimensional Convolutional Neural Networks and Long Short Term Memory networks. This combination is specifically designed to extract spatial features from environmental sensors and analyze temporal sequences without relying on extensive empirical datasets or physical surveys. Through the formulation of the algorithm and the presentation of the core source code the theoretical result demonstrates a lightweight yet robust processing pipeline. This architectural blueprint significantly reduces computational overhead while maintaining high precision in anomaly identification during continuous data streaming. Ultimately the conclusion emphasizes that implementing this conceptual design will profoundly impact modern environmental technology. By providing a scalable software foundation for environmental agencies this framework directly supports the global sustainability agenda specifically aligning with the United Nations Sustainable Development Goals by empowering policymakers with proactive tools to mitigate atmospheric degradation and safeguard communities in rapidly growing urban centers worldwide.
- 31 August 2026
- Published by Sinar Mentari Sundara in Smart Human-centered Emerging Research in Machine Intelligence
- Vol. 2  (2) , 110-120
Open AccessArchitectural Design of a Deep Learning Framework for Real-Time Urban Air Quality Anomaly Detection
The background of this research addresses the escalating environmental crisis caused by rapid urbanization and the urgent need for advanced technological interventions to monitor atmospheric conditions. Rapid industrial growth has led to severe air pollution which directly threatens public health and sustainable development across the globe. Therefore the primary object of this study is the architectural design of a highly efficient predictive framework capable of identifying sudden spikes in air pollution levels across smart city infrastructures. To achieve this the proposed method utilizes a novel deep learning architecture that synergizes one dimensional Convolutional Neural Networks and Long Short Term Memory networks. This combination is specifically designed to extract spatial features from environmental sensors and analyze temporal sequences without relying on extensive empirical datasets or physical surveys. Through the formulation of the algorithm and the presentation of the core source code the theoretical result demonstrates a lightweight yet robust processing pipeline. This architectural blueprint significantly reduces computational overhead while maintaining high precision in anomaly identification during continuous data streaming. Ultimately the conclusion emphasizes that implementing this conceptual design will profoundly impact modern environmental technology. By providing a scalable software foundation for environmental agencies this framework directly supports the global sustainability agenda specifically aligning with the United Nations Sustainable Development Goals by empowering policymakers with proactive tools to mitigate atmospheric degradation and safeguard communities in rapidly growing urban centers worldwide.
- Open Access
- 21 August 2026
- Published by iLearning Journal Center in Jurnal MENTARI: Manajemen, Pendidikan dan Teknologi Informasi
- Vol. 5  (1) , 83-93
Educational Applications and Digital Storytelling for Early Childhood Social Development
The increasing integration of digital technology in early childhood education provides new opportunities to support children’s social development through educational applications and digital storytelling. This study aims to examine how the use of educational applications and digital storytelling influences children’s social development, particularly in communication, cooperation, empathy, and peer interaction. A quantitative approach was employed involving 188 early childhood learners, with data collected through structured questionnaires and observational assessments. The collected data were analyzed using descriptive and inferential statistical techniques to examine the relationship between digital learning media and children’s social development. The findings indicate that educational applications and digital storytelling have a positive impact on children’s social development. Interactive educational activities encourage children to communicate more actively, participate in collaborative tasks, express emotions, and develop empathy toward others. Digital storytelling further supports children’s understanding of social situations by presenting meaningful narratives, characters, and interpersonal experiences that are relevant to their daily lives. These findings suggest that the combination of educational applications and digital storytelling can provide an engaging learning environment that supports the development of essential social skills in early childhood. Therefore, educational applications and digital storytelling can be effectively utilized as complementary learning media when supported by appropriate teacher guidance and balanced with direct social interaction and collaborative activities.
- 21 August 2026
- Published by iLearning Journal Center in Jurnal MENTARI: Manajemen, Pendidikan dan Teknologi Informasi
- Vol. 5  (1) , 83-93
Open AccessEducational Applications and Digital Storytelling for Early Childhood Social Development
The increasing integration of digital technology in early childhood education provides new opportunities to support children’s social development through educational applications and digital storytelling. This study aims to examine how the use of educational applications and digital storytelling influences children’s social development, particularly in communication, cooperation, empathy, and peer interaction. A quantitative approach was employed involving 188 early childhood learners, with data collected through structured questionnaires and observational assessments. The collected data were analyzed using descriptive and inferential statistical techniques to examine the relationship between digital learning media and children’s social development. The findings indicate that educational applications and digital storytelling have a positive impact on children’s social development. Interactive educational activities encourage children to communicate more actively, participate in collaborative tasks, express emotions, and develop empathy toward others. Digital storytelling further supports children’s understanding of social situations by presenting meaningful narratives, characters, and interpersonal experiences that are relevant to their daily lives. These findings suggest that the combination of educational applications and digital storytelling can provide an engaging learning environment that supports the development of essential social skills in early childhood. Therefore, educational applications and digital storytelling can be effectively utilized as complementary learning media when supported by appropriate teacher guidance and balanced with direct social interaction and collaborative activities.
- Open Access
- 10 August 2026
- Published by Sinar Mentari Sundara in Legal Autonomous Web-based Governance (LAW)
- Vol. 1  (2) , 187-199
Legal Governance of Artificial Intelligence in Sustainable Digital Business Ecosystems
The rapid adoption of Artificial Intelligence (AI) within digital business ecosystems has reshaped decision-making, data processing, customer services, and business automation, while simultaneously introducing significant challenges concerning accountability, transparency, data protection, and legal responsibility. This study investigates AI legal governance in sustainable digital business ecosystems by examining how regulatory principles can balance technological innovation with legal certainty, human rights, accountability, and sustainability. A normative legal research approach is employed through statutory, conceptual, and comparative legal analysis. The legal materials include regulations addressing AI-related activities, personal data protection, electronic transactions, digital governance, and relevant international frameworks, supported by scholarly literature and policy documents. The findings indicate that fragmented and technology reactive regulatory approaches are inadequate for addressing the autonomous and continuously evolving characteristics of AI-driven business activities. Effective governance therefore requires an integrated framework incorporating transparency, explainability, human oversight, risk-based regulation, data protection, accountability, and stakeholder responsibility. The proposed framework places human-centered governance at the core of sustainable digital business development by clarifying the responsibilities of AI developers, providers, and business users while reinforcing institutional oversight. This study advances digital legal governance scholarship by connecting AI regulation with sustainable business transformation and responsible technological innovation. Adaptive and accountable legal governance is crucial to ensuring that AI-driven digital ecosystems generate economic value while protecting fundamental rights, public interests, and longterm social sustainability.
- 10 August 2026
- Published by Sinar Mentari Sundara in Legal Autonomous Web-based Governance (LAW)
- Vol. 1  (2) , 187-199
Open AccessLegal Governance of Artificial Intelligence in Sustainable Digital Business Ecosystems
The rapid adoption of Artificial Intelligence (AI) within digital business ecosystems has reshaped decision-making, data processing, customer services, and business automation, while simultaneously introducing significant challenges concerning accountability, transparency, data protection, and legal responsibility. This study investigates AI legal governance in sustainable digital business ecosystems by examining how regulatory principles can balance technological innovation with legal certainty, human rights, accountability, and sustainability. A normative legal research approach is employed through statutory, conceptual, and comparative legal analysis. The legal materials include regulations addressing AI-related activities, personal data protection, electronic transactions, digital governance, and relevant international frameworks, supported by scholarly literature and policy documents. The findings indicate that fragmented and technology reactive regulatory approaches are inadequate for addressing the autonomous and continuously evolving characteristics of AI-driven business activities. Effective governance therefore requires an integrated framework incorporating transparency, explainability, human oversight, risk-based regulation, data protection, accountability, and stakeholder responsibility. The proposed framework places human-centered governance at the core of sustainable digital business development by clarifying the responsibilities of AI developers, providers, and business users while reinforcing institutional oversight. This study advances digital legal governance scholarship by connecting AI regulation with sustainable business transformation and responsible technological innovation. Adaptive and accountable legal governance is crucial to ensuring that AI-driven digital ecosystems generate economic value while protecting fundamental rights, public interests, and longterm social sustainability.
- Open Access
- 29 June 2026
- Published by iLearning Journal Center in Startupreneur Business Digital (SABDA Journal)
- Vol. 5  (2) , 119-130
Optimizing the SOR Method for Purchasing Decisions of Fashion Startupreneur on the Shopee Live Feature
The rapid growth of social commerce in Indonesia has opened wide opportu- nities for fashion business owners to leverage social commerce tools built into e-commerce platforms as a means of steering how consumers reach their buying choices. This study aims to examine how three core dimensions of social commerce interactivity, social presence, and informativeness affect fashion purchase decisions on the Shopee platform, using perceived value and emotional trust as mediating factors under the Stimulus Organism Response (SOR) framework. A further aim is to determine whether cognitive or affective processes play a greater role in shaping consumers’ decisions to buy fashion items online. Data were collected from Indonesian Shopee users who had previously bought fashion products from stores run by startup founders, and the data were then analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). The findings supported all of the proposed relationships. Both informativeness (β = 0.553) and interactivity (β = 0.383) showed a positive impact on perceived value, whereas social presence (β = 0.677) had a substantial positive effect on emotional trust. In turn, perceived value (β = 0.506) and emotional trust (β = 0.202) each positively influenced purchase decisions. The proposed model accounted for 74.7% of the variance in perceived value and 45.6% of the variance in purchase decisions, indicating that the cognitive route via perceived value carried more weight than the affective route via emotional trust in driving online fashion purchases. Moreover, informativeness emerged as the most impactful social commerce factor in boosting both perceived value and purchase intention.
- 29 June 2026
- Published by iLearning Journal Center in Startupreneur Business Digital (SABDA Journal)
- Vol. 5  (2) , 119-130
Open AccessOptimizing the SOR Method for Purchasing Decisions of Fashion Startupreneur on the Shopee Live Feature
The rapid growth of social commerce in Indonesia has opened wide opportu- nities for fashion business owners to leverage social commerce tools built into e-commerce platforms as a means of steering how consumers reach their buying choices. This study aims to examine how three core dimensions of social commerce interactivity, social presence, and informativeness affect fashion purchase decisions on the Shopee platform, using perceived value and emotional trust as mediating factors under the Stimulus Organism Response (SOR) framework. A further aim is to determine whether cognitive or affective processes play a greater role in shaping consumers’ decisions to buy fashion items online. Data were collected from Indonesian Shopee users who had previously bought fashion products from stores run by startup founders, and the data were then analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). The findings supported all of the proposed relationships. Both informativeness (β = 0.553) and interactivity (β = 0.383) showed a positive impact on perceived value, whereas social presence (β = 0.677) had a substantial positive effect on emotional trust. In turn, perceived value (β = 0.506) and emotional trust (β = 0.202) each positively influenced purchase decisions. The proposed model accounted for 74.7% of the variance in perceived value and 45.6% of the variance in purchase decisions, indicating that the cognitive route via perceived value carried more weight than the affective route via emotional trust in driving online fashion purchases. Moreover, informativeness emerged as the most impactful social commerce factor in boosting both perceived value and purchase intention.
- Open Access
- 26 May 2026
- Published by Sinar Mentari Sundara in Health, Empathy, and AI Learning (HEAL)
- Vol. 1  (2) , 102-112
An Exhaustive Survey to Understand Music Generation based on Artificial Intelligence
Music generation is a creative process that requires the ability to understand melody, rhythm, harmony, structure, and emotional expression. Although music has long been viewed as a human-centered artistic domain, the development of Artificial Intelligence has opened new opportunities for automatic music generation with minimal human intervention. This study aims to analyze the use of AI algorithms in music generation, identify commonly used approaches, and examine existing gaps in producing high-quality musical compositions. The study reviews several AI-based methods, including recurrent neural networks, Deep Composer, WaveNet, memetic algorithms, generative adversarial networks, reinforcement learning, and transformer-based models. In addition, publicly available music datasets and GAN-based synthetic data generation are considered to support the training process. The findings indicate that deep learning models are effective in learning musical patterns, while reinforcement learning and transformer-based preprocessing can improve sequence understanding, adaptability, and structural coherence. However, current models still face challenges in duplicating specific artist styles, maintaining complete song structure, and generating music with strong uniqueness. Therefore, integrating deep learning, reinforcement learning, GAN-based synthetic data, and transformer preprocessing offers a promising direction for improving AI-generated music quality and supporting future research in automatic music composition.
- 26 May 2026
- Published by Sinar Mentari Sundara in Health, Empathy, and AI Learning (HEAL)
- Vol. 1  (2) , 102-112
Open AccessAn Exhaustive Survey to Understand Music Generation based on Artificial Intelligence
Music generation is a creative process that requires the ability to understand melody, rhythm, harmony, structure, and emotional expression. Although music has long been viewed as a human-centered artistic domain, the development of Artificial Intelligence has opened new opportunities for automatic music generation with minimal human intervention. This study aims to analyze the use of AI algorithms in music generation, identify commonly used approaches, and examine existing gaps in producing high-quality musical compositions. The study reviews several AI-based methods, including recurrent neural networks, Deep Composer, WaveNet, memetic algorithms, generative adversarial networks, reinforcement learning, and transformer-based models. In addition, publicly available music datasets and GAN-based synthetic data generation are considered to support the training process. The findings indicate that deep learning models are effective in learning musical patterns, while reinforcement learning and transformer-based preprocessing can improve sequence understanding, adaptability, and structural coherence. However, current models still face challenges in duplicating specific artist styles, maintaining complete song structure, and generating music with strong uniqueness. Therefore, integrating deep learning, reinforcement learning, GAN-based synthetic data, and transformer preprocessing offers a promising direction for improving AI-generated music quality and supporting future research in automatic music composition.
- Open Access
- 26 May 2026
- Published by Pandawan in ADI Bisnis Digital Interdisiplin Jurnal
- Vol. 7  (1) , 82-95
Optimizing Digital Marketing for a Healthy Snack Business Through Collaborative Instagram Posts
Increasing awareness of children’s health has shifted parents’ preferences from conventional snacks to healthier alternatives. The primary issue is that parents’ purchasing decisions are strongly influenced by trust and community validation. Social media has transformed from a platform for sharing information into an instrument of social validation that helps parents select products for their children. As an adaptive platform, Instagram introduced the collaboration post feature, enabling two accounts to jointly publish content. The urgency of this research lies in the challenge faced by brands in determining whether independently posting content is sufficiently effective or utilizing collaboration features is necessary to penetrate platform algorithms and reach audiences. This study aims to analyze the significant differences in Engagement Rate (ER) between Regular Post and Collaboration Post involving credible third parties, such as pediatricians, nutritionists, and mom influencers. The novelty of this research lies in optimizing the differences in the strategic functions of engagement and reach within collaborative social media marketing strategies. The analysis results indicate that regular posts are more effective in fostering community engagement, whereas collaboration posts are more effective in expanding audience reach. This finding is supported by the Independent Sample t-test, indicating a significant difference in engagement rates between regular posts and collaboration posts (t = 2.897; p = 0.010 < 0.05). Therefore, this study contributes to understanding the strategic trade-off between audience reach expansion and audience engagement levels in digital marketing strategies
- 26 May 2026
- Published by Pandawan in ADI Bisnis Digital Interdisiplin Jurnal
- Vol. 7  (1) , 82-95
Open AccessOptimizing Digital Marketing for a Healthy Snack Business Through Collaborative Instagram Posts
Increasing awareness of children’s health has shifted parents’ preferences from conventional snacks to healthier alternatives. The primary issue is that parents’ purchasing decisions are strongly influenced by trust and community validation. Social media has transformed from a platform for sharing information into an instrument of social validation that helps parents select products for their children. As an adaptive platform, Instagram introduced the collaboration post feature, enabling two accounts to jointly publish content. The urgency of this research lies in the challenge faced by brands in determining whether independently posting content is sufficiently effective or utilizing collaboration features is necessary to penetrate platform algorithms and reach audiences. This study aims to analyze the significant differences in Engagement Rate (ER) between Regular Post and Collaboration Post involving credible third parties, such as pediatricians, nutritionists, and mom influencers. The novelty of this research lies in optimizing the differences in the strategic functions of engagement and reach within collaborative social media marketing strategies. The analysis results indicate that regular posts are more effective in fostering community engagement, whereas collaboration posts are more effective in expanding audience reach. This finding is supported by the Independent Sample t-test, indicating a significant difference in engagement rates between regular posts and collaboration posts (t = 2.897; p = 0.010 < 0.05). Therefore, this study contributes to understanding the strategic trade-off between audience reach expansion and audience engagement levels in digital marketing strategies
- 30 April 2026
- Published by Pandawan in ADI Pengabdian Kepada Masyarakat
- Vol. 6  (2) , 186-198
Implementation of Local SEO on Google Maps for Increasing Footfall Conversion in Building Store Business
In today’s digital era, consumer behavior in searching for building materials and home renovation services has shifted to digital platforms, especially Google Maps. Traditional building material businesses still lag behind in digital visibility and often lose competitiveness against modern building depots. This community service activity aims to help traditional MSMEs survive and grow through digitalization by increasing business visibility and customer footfall conversion. This study applied a local Search Engine Optimization (SEO) approach through a case study of a building material MSME. The research stages included initial observation of the business profile and customer visits, implementation of Local SEO optimization strategies, and final observation to evaluate the resulting changes. Optimization strategies included completing business information, applying local keywords, adding visual content, and actively managing customer reviews. The novelty of this research lies in measuring footfall conversion through the implementation of Local SEO on Google Maps for traditional building material MSMEs. The study focuses not only on digital business profile management but also on the direct impact on increasing physical customer visits. The findings indicate an increase in local search visibility and a 76.68% rise in physical customer visits after optimization. Utilizing free Google Maps features proved to be an efficient and practical digitalization strategy to improve MSME competitiveness and business sustainability.
- 30 April 2026
- Published by Pandawan in ADI Pengabdian Kepada Masyarakat
- Vol. 6  (2) , 186-198
Implementation of Local SEO on Google Maps for Increasing Footfall Conversion in Building Store Business
In today’s digital era, consumer behavior in searching for building materials and home renovation services has shifted to digital platforms, especially Google Maps. Traditional building material businesses still lag behind in digital visibility and often lose competitiveness against modern building depots. This community service activity aims to help traditional MSMEs survive and grow through digitalization by increasing business visibility and customer footfall conversion. This study applied a local Search Engine Optimization (SEO) approach through a case study of a building material MSME. The research stages included initial observation of the business profile and customer visits, implementation of Local SEO optimization strategies, and final observation to evaluate the resulting changes. Optimization strategies included completing business information, applying local keywords, adding visual content, and actively managing customer reviews. The novelty of this research lies in measuring footfall conversion through the implementation of Local SEO on Google Maps for traditional building material MSMEs. The study focuses not only on digital business profile management but also on the direct impact on increasing physical customer visits. The findings indicate an increase in local search visibility and a 76.68% rise in physical customer visits after optimization. Utilizing free Google Maps features proved to be an efficient and practical digitalization strategy to improve MSME competitiveness and business sustainability.
- Open Access
- 21 April 2026
- Published by Sinar Mentari Sundara in Journal of Orange Technology
- Vol. 3  (1) , 37-48
Human Centered Artificial Intelligence Through GDPR Compliant Data Science Workflow Implementation
The rapid advancement of Artificial Intelligence (AI) and data science technologies has transformed organizational decision-making across healthcare, finance, and digital services. Despite these advancements, the increasing use of personal data has intensified concerns regarding privacy, transparency, ac countability, and public trust. Building trustworthy Human-Centered AI requires data science workflows that integrate regulatory compliance, ethical principles, and responsible governance throughout the AI lifecycle. This study aims to examine how GDPR-compliant data science workflow implementation supports the development of trustworthy Human-Centered AI through a qualitative and practice-oriented research approach. The study synthesizes evidence from recent literature, industry case studies, expert perspectives, and governance oriented analytical frameworks to identify effective strategies for integrating privacy by design, data minimization, transparency, accountability, and privacy-preserving techniques, including anonymization, pseudonymization, and differential privacy, into data science workflows. The findings indicate that successful implementation depends not only on technical safeguards but also on strong organizational governance, continuous compliance monitoring, and cross-functional collaboration among legal, technical, and managerial stakeholders. Furthermore, the integration of explainable and governance-aware machine learning models improves transparency, strengthens stakeholder trust, and supports responsible human-centered AI without significantly reducing analytical performance. This study proposes a structured GDPR-compliant data science workflow framework that enables organizations to balance analytical effectiveness, regulatory compliance, and human-centered principles while fostering trustworthy, transparent, and sustainable Artificial Intelligence for real-world digital innovation.
- 21 April 2026
- Published by Sinar Mentari Sundara in Journal of Orange Technology
- Vol. 3  (1) , 37-48
Open AccessHuman Centered Artificial Intelligence Through GDPR Compliant Data Science Workflow Implementation
The rapid advancement of Artificial Intelligence (AI) and data science technologies has transformed organizational decision-making across healthcare, finance, and digital services. Despite these advancements, the increasing use of personal data has intensified concerns regarding privacy, transparency, ac countability, and public trust. Building trustworthy Human-Centered AI requires data science workflows that integrate regulatory compliance, ethical principles, and responsible governance throughout the AI lifecycle. This study aims to examine how GDPR-compliant data science workflow implementation supports the development of trustworthy Human-Centered AI through a qualitative and practice-oriented research approach. The study synthesizes evidence from recent literature, industry case studies, expert perspectives, and governance oriented analytical frameworks to identify effective strategies for integrating privacy by design, data minimization, transparency, accountability, and privacy-preserving techniques, including anonymization, pseudonymization, and differential privacy, into data science workflows. The findings indicate that successful implementation depends not only on technical safeguards but also on strong organizational governance, continuous compliance monitoring, and cross-functional collaboration among legal, technical, and managerial stakeholders. Furthermore, the integration of explainable and governance-aware machine learning models improves transparency, strengthens stakeholder trust, and supports responsible human-centered AI without significantly reducing analytical performance. This study proposes a structured GDPR-compliant data science workflow framework that enables organizations to balance analytical effectiveness, regulatory compliance, and human-centered principles while fostering trustworthy, transparent, and sustainable Artificial Intelligence for real-world digital innovation.