Claim Missing Document
Check
Articles

Found 21 Documents
Search

Machine Learning-Based Multi-Sensor IoT System for Intelligent Indoor Fire Detection Anggy Pradifta Junfithranaa; Hadi Almohab; Deshinta Arrova Dewi
Journal of Educational Technology and Learning Creativity Vol. 4 No. 1 (2026): June
Publisher : Cahaya Ilmu Cendekia Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37251/jetlc.v4i1.2614

Abstract

Purpose of the study: This study aims to develop an intelligent indoor fire detection system by integrating low-cost Internet of Things (IoT) sensors with machine learning-based multi-sensor data fusion to improve early fire hazard detection accuracy while reducing false alarms compared to conventional single-sensor fire detection systems. Methodology: The system is implemented using an ESP32 microcontroller connected to temperature, humidity, flame, and sound sensors for real-time data acquisition. A dataset of 1,500 sensor samples is collected and labeled into Normal, Fire-Risk, and Fire classes. Decision Tree, Support Vector Machine, and Random Forest classifiers are trained and evaluated using Python-based machine learning libraries. Main Findings: Experimental results indicate that the Random Forest model outperforms the other classifiers, achieving 95% overall accuracy, perfect recall for fire events, and a Macro ROC-AUC score of 0.993. Feature importance analysis reveals that humidity and temperature are the most influential parameters for early fire detection in indoor environments. Novelty/Originality of this study: This study proposes a lightweight intelligent fire detection framework that integrates multi-sensor Internet of Things data including temperature, humidity, flame, and sound signals with machine learning–based classification for indoor environments. Unlike conventional systems that rely on single-sensor or threshold-based detection, the proposed approach utilizes multi-sensor data fusion and ensemble learning to improve early fire-risk identification while remaining computationally efficient for low-cost platforms such as the ESP32 microcontroller.
Integrated Virtual Reality Learning Framework with Digital Ecosystem for Enhancing Physics Conceptual Understanding Eko Risdianto; Joselin Santos; Noel Lomerio; Deshinta Arrova Dewi; M. Esad Kuloglu; Sultan Hammad Alshammari; Ressti Nurfitriani
Journal of Educational Technology and Learning Creativity Vol. 4 No. 1 (2026): June
Publisher : Cahaya Ilmu Cendekia Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37251/jetlc.v4i1.2955

Abstract

Purpose of the study: This study develops and evaluates an integrated virtual reality learning framework to improve high school students’ physics conceptual understanding. The framework combines immersive 360° virtual reality videos via Kuula with Google Classroom, ClassPoint, digital flipbooks, and PhET simulations within a problem-based learning environment. Methodology: This study employed a research and development approach using the ADDIE model, combined with a pre-experimental one-group pretest–posttest design. The participants consisted of three cohorts of high school students (n = 115) across different physics topics: Kinematics (n = 36), fluids (n = 39), and particle dynamics (n = 40). The framework was validated by three experts using structured instruments assessing content, media, language, and presentation aspects. Data were collected through validation sheets, student response questionnaires, and conceptual understanding tests. Data analysis included percentage-based measures, N-Gain, Shapiro Wilk tests, paired t-tests, Wilcoxon signed rank tests, and effect size calculations. Main Findings: The results indicate high validity (88.89%–97%) and practicality (86%–98%). The implementation was associated with significant improvements in conceptual understanding, reflected in high N-Gain scores (0.81–0.87) and large effect sizes (p < 0.05). These findings suggest that the integrated virtual reality based learning ecosystem can effectively support conceptual understanding within the studied context. Novelty/Originality of this study: The novelty lies in the systematic integration of the Kuula platform within a multi-component digital learning ecosystem under a problem based learning framework, as well as its application across multiple physics topics to demonstrate consistent learning outcomes.
Robotics Workshop to Increase Motivation and Science Literacy of SMP IT Khairu Ummah Rejang Lebong Students Afrizal Mayub; M. Lutfi Firdaus; Fahmizal Fahmizal; Aceng Ruyani; Lazfihma Lazfihma; Deshinta Arrova Dewi
Aktual: Jurnal Pengabdian Kepada Masyarakat Vol. 3 No. 1 (2025): Aktual: Jurnal Pengabdian Kepada Masyarakat January 2025
Publisher : CV Media Inti Teknologi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58723/aktual.v3i1.347

Abstract

This activity aims to Increase the Motivation and Science Literacy of SMP IT Khairu Ummah Rejang Lebong Students. This activity is one form of implementing the tridharma of higher education. One of the tri dharmas of higher education is community service, this service activity is an activity that every lecturer must carry out, therefore lecturers in the Master of Science program at Bengkulu University collaborated with IT SMP teacher Khairu Ummah Rejang Lebong Bengkulu on September 28, 2024, to carry out service activities at SMP IT Khairu Ummah Rejang Lebong. This collaboration takes the form of community service activities at SMP IT Khairu Ummah Rejang Lebong Bengkulu with the title "Robotics Workshop to Increase Motivation and Science Literacy of SMP IT Khairu Ummah Rejang Lebong Students". From these activities, it can be concluded that the teaching material delivered at the Robotics Workshop uses interactive multimedia which includes lectures, virtual demos, animations, visualizations, simulations and videos which have proven successful in motivating students in the motivated category with a score of 3.84 and increasing scientific literacy in the high category with a score of 3.93.
Comparative study of NBC, SVM, and CNN performance in sentiment analysis of the millennial farmers program on platform X Adamu Abu Bakar Ibrahim; Junaedi Rahmat; Deshinta Arrova Dewi
Indonesian Journal of Machine Learning and Intelligent Systems Vol. 1 No. 1 (2026)
Publisher : Indonesian Artificial Neural Network Society

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Public perception of government programs is often expressed on social media, making sentiment analysis a vital tool for understanding public opinion. This research presents a comparative study of three machine learning algorithms, Naive Bayes Classifier (NBC), Support Vector Machine (SVM) and Convolutional Neural Network (CNN), applied to sentiment analysis on the Millennial Farmers Program 2024 from Platform X. The study aims to measure the performance of each algorithm in terms of accuracy, execution time, and memory efficiency by adopted a quantitative approach with a comparative experimental design. Sentiment data were collected from platform X (Twitter) between October 2024 and January 2025, with a total of 5,177 raw tweets acquired using the keyword "petani_milenial". After data preprocessing steps, including cleaning, case folding, tokenization, stopword removal, and stemming, the data's sentiment (positive, negative, neutral) was automatically labeled using the IndoBERT model. The sentiment analysis results revealed a dominance of neutral sentiment (832 tweets), followed by negative (331 tweets) and positive (197 tweets). Model performance evaluation showed that SVM achieved the highest overall accuracy at 79%, demonstrating superior capability in classifying neutral sentiment. CNN, with an accuracy of 74.26%, stood out in recall for positive sentiment (73%), making it an effective choice for comprehensive identification of positive sentiment. Meanwhile, NBC, with 78% accuracy, proved to be the most efficient in terms of computation time and memory usage. The study concludes that the optimal model selection greatly depends on the specific use-case priority, whether overall accuracy, positive sentiment identification, or computational efficiency. The results show that CNN outperforms NBC and SVM in accuracy, while NBC is the fastest and consumes the least memory. These findings can help inform future research and implementation in large-scale sentiment analysis applications.
Strategic ESG-Driven Human Resource Practices: Transforming Employee Management for Sustainable Organizational Growth Darul Wiyono; Deshinta Arrova Dewi; Ema Ambiapuri; Nur Aini Parwitasari; Deni Supardi Hambali
Jurnal Organisasi dan Manajemen Vol. 21 No. 1 (2025)
Publisher : LPPM Universitas Terbuka

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33830/jom.v21i1.9786.2025

Abstract

Purpose - This research explores the impact of Environmental, Social, and Governance (ESG) practices on employee performance and well-being in private higher education institutions in Bandung, West Java. It seeks to provide insights into how effective ESG integration can enhance organizational performance and employee satisfaction. Methodology - A quantitative approach employing Partial Least Squares Structural Equation Modeling (PLS-SEM) analyzed the relationships among constructs: environmental practices, social practices, governance practices, and the dependent variable, employee performance and well-being. Data were collected from 270 respondents through stratified random sampling across various administrative roles in 138 private higher education institutions. Findings- The results showed that environmental practices positively impact employee performance and well-being, with social practices also contributing. Governance practices mediate and amplify these effects. These findings emphasize the importance of integrating sustainable practices into organizational strategies to improve employee outcomes and overall institutional performance. Originality—This study explores the impact of ESG practices on employee performance in private higher education institutions, focusing on sustainability to enhance engagement and productivity. Unlike previous research, which focused primarily on corporate sectors or public universities, it emphasizes the unique challenges of private institutions, particularly in Bandung, offering new insights into governance practices as mediators.
Performance Evaluation of a Rabbit Manure-Based Biogas Power System: Slurry Dynamics, Energy Yield, and Conversion Efficiency As'ad Shidqy Aziz; Deshinta Arrova Dewi; Daeng Rahmatullah; Muhammad ‘Izzuddin Al-Qassam; Ayusta Lukita Wardani; Fithrotul Irda Amaliah; Onny Setyawati
Buletin Ilmiah Sarjana Teknik Elektro Vol. 8 No. 4 (2026): August
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/biste.v8i4.16870

Abstract

The increasing demand for sustainable energy has encouraged decentralized biogas-based power systems, yet a critical research gap remains regarding their field-scale integration and multi-parameter thermodynamic evaluations under real farming conditions. The research contribution is the field-scale operational integration and continuous performance evaluation of a 500 L rabbit manure biodigester coupled with a three-stage purification unit and a modified 1000 W generator set. Utilizing a transparent, reproducible mathematical framework, fresh rabbit manure was digested under a 37-day hydraulic retention time. Results revealed that the accumulated slurry occupied 35.5% of the biodigester volume, leaving 64.5% available as headspace for passive thermodynamic pressure management. The purified biogas successfully operated a 40 W barn lighting load for 12 h day⁻¹, generating a stable average electrical energy output of 0.485 kWh day⁻¹ and a specific energy yield of 0.202 kWh kg⁻¹ of fresh manure. The integrated system achieved a validated Specific Energy Consumption (SEC) value of 0.99 and an overall energy conversion efficiency of 64.7%. While this investigation is limited by its small-scale setup and a 31-day batch cycle, the practical implications demonstrate that this layout provides a viable, standalone template for circular waste management and rural energy independence, establishing an empirical baseline to motivate future automated or upscaled microgrid architectures.
Utilizing Sentiment Analysis for Reflect and Improve Education in Indonesia Henderi Henderi; Asro Asro; Agus Sulaiman; Tri Basuki Kurniawan; Deshinta Arrova Dewi; Mashal AlQudah
Journal of Applied Data Sciences Vol 6, No 1: JANUARY 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i1.527

Abstract

This study explores the potential of sentiment analysis in providing valuable insights into education in Indonesia based on comments from the YouTube platform. Utilizing the Naive Bayes Classifier method, this research analyzed 13,386 processed comments out of 17,920 original comments. The results show that 53.8% of comments were negative, while 28.5% were positive, and 17.7% were neutral, reflecting diverse perspectives on existing educational issues. The Accuracy of this model reached up to 72.51% with testing on various sample sizes (10%-30%), indicating the model's effectiveness in identifying sentiments. Although the model tends to classify comments as unfavorable, this opens opportunities for introspection and improvement within the educational system. Further analysis with a word cloud revealed dominant keywords, indicating areas that require more attention in public discussions about education. By leveraging this sentiment analysis, the study offers practical and valuable guidance for policymakers to reflect on and enhance educational strategies and policies in Indonesia. This research measures public reactions and aims to foster more constructive and inclusive discussions about the sustainable development of education in Indonesia.
Leveraging Data Analytics for Student Grade Prediction: A Comparative Study of Data Features Misinem Misinem; Tri Basuki Kurniawan; Deshinta Arrova Dewi; Mohd Zaki Zakaria; Che Mohd Alif Nazmi
Journal of Applied Data Sciences Vol 5, No 4: DECEMBER 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i4.442

Abstract

In educational settings, a persistent challenge lies in accurately identifying and supporting students at risk of underperformance or grade retention. Traditional approaches often fall short by applying generalized interventions that fail to address specific academic needs, leading to ineffective outcomes and increased grade repetition. This study advocates for integrating machine learning algorithms into educational assessment practices to address these limitations. By leveraging historical and current performance data, machine learning models can help identify students needing additional support early in their academic journey, allowing for precise and timely interventions. This research examines the effectiveness of three machine learning algorithms: Naive Bayes, Deep Learning, and Decision Trees. Naive Bayes, known for its simplicity and efficiency, is well-suited for initial data screening. Deep Learning excels at uncovering complex patterns in large datasets, making it ideal for nuanced predictions. Decision Trees, with their interpretable and actionable outputs, provide clear decision paths, making them particularly advantageous for educational applications. Among the models tested, the Decision Tree algorithm demonstrated the highest performance, achieving an accuracy rate of 86.68%. This high precision underscores its suitability for educational contexts where decisions need to be based on reliable, interpretable data. The results strongly support the broader application of Decision Tree analysis in educational practices. By implementing this model, educational administrators can better identify at-risk students, tailor interventions to meet individual needs, and ultimately improve student success rates. This study suggests that Decision Trees could become a vital tool in data-driven strategies to enhance student retention and optimize academic outcomes.
Scalable Machine Learning Approaches for Real-Time Anomaly and Outlier Detection in Streaming Environments Deshinta Arrova Dewi; Harprith Kaur Rajinder Singh; Jeyarani Periasamy; Tri Basuki Kurniawan; Henderi Henderi; M. Said Hasibuan
Journal of Applied Data Sciences Vol 5, No 4: DECEMBER 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i4.444

Abstract

The prevalence of streaming data across various sectors poses significant challenges for real-time anomaly detection due to its volume, velocity, and variability. Traditional data processing methods often need to be improved for such dynamic environments, necessitating robust, scalable, and efficient real-time analysis systems. This study compares two advanced machine learning approaches—LSTM autoencoders and Matrix Profile algorithms—to identify the most effective method for anomaly detection in streaming environments using the NYC taxi dataset. Existing literature on anomaly detection in streaming data highlights various methodologies, including statistical tests, window-based techniques, and machine learning models. Traditional methods like the Generalized ESD test have been adapted for streaming data but often require a full historical dataset to function effectively. In contrast, machine learning approaches, particularly those using LSTM networks, are noted for their ability to learn complex patterns and dependencies, offering promising results in real-time applications. In a comparative analysis, LSTM autoencoders significantly outperformed other methods, achieving an F1-score of 0.22 for anomaly detection, notably higher than other techniques. This model demonstrated superior capability in capturing temporal dependencies and complex data patterns, making it highly effective for the dynamic and varied data in the NYC taxi dataset. The LSTM autoencoder's advanced pattern recognition and anomaly detection capabilities confirm its suitability for complex, high-velocity streaming data environments. Future research should explore the integration of LSTM autoencoders with other machine-learning techniques to enhance further the accuracy, scalability, and efficiency of anomaly detection systems. This study advances our understanding of scalable machine-learning approaches and underscores the critical importance of selecting appropriate models based on the specific characteristics and challenges of the data involved.
Efficient Fruit Grading and Selection System Leveraging Computer Vision and Machine Learning Deshinta Arrova Dewi; Tri Basuki Kurniawan; Rajermani Thinakaran; Malathy Batumalay; Shabana Habib; Muhammad Islam
Journal of Applied Data Sciences Vol 5, No 4: DECEMBER 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i4.443

Abstract

Automated fruit grading is crucial to overcoming the time and accuracy challenges posed by manual methods, which are often limited by subjective human judgment. This study introduces an intelligent grading system leveraging computer vision and AI to improve speed and consistency in assessing fruit quality. Using high-resolution imaging and advanced feature extraction, including grayscale processing, binarization, and enhancement, the system achieves non-destructive, efficient sorting for fruits like apples, bananas, and oranges. Grayscale processing reduces image complexity while preserving essential details, binarization isolates the fruit from its background, and enhancement highlights critical features. Notably, the Edge Pixel method proved most effective, achieving 79.20% accuracy in grading, while the Grayscale Pixel method reached 93.94% accuracy for fruit types. Edge Pixel also achieved 80.32% in differentiating grading types, showcasing its ability to capture essential shapes and edges. Fruits are classified into four grades: Grade_01 (highest quality), Grade_02 (minor imperfections), Grade_03 (notable defects but consumable), and Grade_04 (unfit for consumption). A specialized dataset supports model training, ensuring practical real-world application. The study concludes that this automated system offers significant improvements over traditional grading, providing a scalable, objective, and reliable solution for the agricultural sector, ultimately enhancing productivity and quality assurance.
Co-Authors Abd Muin Abhijit Banubakode ACENG RUYANI Adamu Abu Bakar Ibrahim Afrizal Mayub Agus Sulaiman Ahmad Bahrudin Alfarid Hendro Yuwono Aliyu Sabo Anggy Pradifta Junfithranaa Asro Asro Ayusta Lukita Wardani Aziz, As’ad Shidqy Cecep Soleh Kurniawan Che Mohd Alif Nazmi Daeng Rahmatullah Dani Durahman Darul Wiyono Deni Supardi Hambali Dhendra Marutho Eko Risdianto Ema Ambiapuri Eva Yulia Puspaningrum Fahmizal Fahmizal Fithrotul Irda Amaliah Giraldo Stevanus Hadi Almohab Hajjar Ayu Cahyani Kuswardhani Harprith Kaur Rajinder Singh Harsha J. Sarode Henderi . I Gede Susrama Mas Diyasa I Gede Susrama Mas Diyasa Jeyarani Periasamy Joselin Santos JosephNg Poh Soon Junaedi Rahmat Lazfihma Lazfihma Liton Chandra Paul M. Esad Kuloğlu M. Lutfi Firdaus M. Said Hasibuan M. Tohir Mahesh Shankarrao Patil Malathy Batumalay Malathy Batumalay Margareta Valencia Suci Handayani Mashal AlQudah Md. Abdul Kader Jilani Md. Ashraful Haque Md. Sharif Ahammed Misinem Misinem Mohammad Idhom Mohd Zaki Zakaria Muhammad Islam Muhammad Nasir Muhammad ‘Izzuddin Al-Qassam Muljono, - Nafi Isbadrianingtyas Narinderjit Singh Sawaran Singh Noel Lomerio Nur Aini Parwitasari Nurul Shaira Binti Bujang Nutan Patil Onny Setyawati Pangestu Sandya Etniko Siagian Panji Adam Agus Putra Prakash Tukaram Patil Pramudita, Yoga Dwitya Prismahardi Aji Riyantoko Raden Arief Nugroho Rajani Balakrishnan Rajermani Thinakaran Rajermani Thinakaran Redwan A. Ananta Ressti Nurfitriani Reza Rahmadian Ruri Suko Basuki Sayyidah Humairah Shabana Habib Suhardini, Eni Dasuki Sultan Hammad Alshammari Tri Basuki Kurniawan Tri Basuki Kurniawan Tri Basuki Kurniawan Tri Basuki Kurniawan Ugochi Okengwu Vijayakumar Varadarajan Vugar Hacimahmud Abdullayev Wahyudi Agustiono Wahyudi Setiawan Wan Suryani Wan Awang Widi Aribowo Yo Ceng Giap Zakka Ugih Rizqi