cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
-
Contact Email
edukomputika@mail.unnes.ac.id
Phone
-
Journal Mail Official
edukomputika@mail.unnes.ac.id
Editorial Address
Sekaran, Kec. Gn. Pati, Kota Semarang, Jawa Tengah 50229
Location
Kota semarang,
Jawa tengah
INDONESIA
Edu Komputika Journal
Published by Universitas Negeri Semarang
ISSN : -     EISSN : 2599297X     DOI : https://doi.org/10.15294/edukom
Core Subject : Education,
Education
Edu Komputika Journal uses Open Journal Systems (OJS) for online journal management in submission, review, copyediting, and publication. Submitted manuscripts are written in English and should follow the style of the Edu Komputika Journal. Manuscripts are original research results, or theoretical/literature study results that have never been published in other journals or are not considered for publication elsewhere. The author should follow all the provisions and processes. Accepted papers will be available online and will be charged a publication fee.
Arjuna Subject : Ilmu Sosial - Pendidikan
Articles 45 Documents
 1   2   3   4   5 
Ensemble RNN–Random Forest Model for Earthquake Prediction Based onSpatio-Temporal Seismic Data Meyn Choudy Riovan Kaotel; Gladly Caren Rorimpandey; Sondy Campvid Kumajas; Muhammad Zulkifli
Edu Komputika Journal Vol. 12 No. 1 (2025): Edu Komputika Journal
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/edukom.v12i1.35332

Abstract

This study proposes a hybrid Ensemble RNN–Random Forest (RNN–RF) model for short-term earthquake prediction based on spatio-temporal seismic data from the Sulawesi–Maluku region. The purpose of this research is to develop a lightweight and interpretable machine learning framework that integrates temporal and spatial features using local datasets provided by the Manado Geophysical Station of BMKG. The methodology includes six stages: data acquisition, preprocessing, feature engineering, model development, ensemble integration, and evaluation. The RNN captures sequential dependencies in seismic activity, while the Random Forest learns spatial and contextual relationships such as fault proximity and event clustering. The ensemble fuses probabilistic outputs (0.75 RNN and 0.25 RF) followed by domain-based calibration using mean magnitude, event frequency, and fault distance. Experimental results show that the proposed ensemble achieved F1 = 0.89 and AUC = 0.975, outperforming individual RNN and RF models in predictive stability and accuracy. The model demonstrates that integrating domain-specific adjustments enhances both recall and precision, while maintaining interpretability for operational deployment. This study contributes to explainable AI in seismology by bridging deep temporal modeling with geophysical reasoning, offering a scalable approach for early-warning applications in Indonesia.
Usability Scale to Measure the Acceptability of Interactive Learning Media for Ratio Material in Grade VI Diki Somantri; Yunus Abidin; Dede Tri Kurniawan; Dinie Anggraeni Dewi
Edu Komputika Journal Vol. 12 No. 2 (2025): Edu Komputika Journal
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/edukom.v12i2.35761

Abstract

Mathematics learning, particularly for abstract topics such as ratios, requires a transition from additive to multiplicative thinking, which is often challenging to visualize. To support teachers, an Interactive Learning Media (ILM-Ratio) was developed using Canva to ensure accessibility and ease of content modification. Before measuring student effectiveness, it is crucial to assess the media's technical acceptability from the teachers' perspective, as they are the primary gatekeepers of classroom technology. This study aims to evaluate the usability of ILM-Ratio using the System Usability Scale (SUS). A total of 47 sixth-grade teachers participated by simulating a teaching session using the media. The SUS calculation yielded an average score of 90 (Grade A), categorized as "Excellent." This score indicates that ILM-Ratio possesses high usability and imposes minimal technical burden on teachers. These findings suggest that the media is technically robust and ready for classroom implementation to support the visualization of ratio concepts.
Interactive Multimedia as a More Effective and Engaging Solution for Visual Programming Learning Achmad Darojat; Armedo Futhi Munadzar
Edu Komputika Journal Vol. 12 No. 2 (2025): Edu Komputika Journal
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/edukom.v12i2.36282

Abstract

This study developed and evaluated an interactive multimedia learning product for the Visual Programming course at Sekolah Tinggi Teknologi STIKMA Internasional. Using the ADDIE development model Analysis, Design, Development, Implementation, and Evaluation the research aimed to improve learning effectiveness and student engagement through multimedia-based instruction. The product, designed as an executable (EXE) application, includes six chapters integrating text, animation, video, quizzes, and interactive navigation. Validation involved four experts in content, media, language, and user interface, as well as student trials in both small-group and field settings. Expert validation results averaged 89.26%, categorized as highly valid, while student trials showed an average of 84.8%. Field testing indicated that all students achieved the course passing standard, with scores ranging from 60 to 100. These findings demonstrate that the interactive multimedia effectively enhances students’ comprehension, motivation, and learning outcomes. The results are consistent with the Cognitive Theory of Multimedia Learning, which emphasizes the integration of visual and verbal channels to optimize cognitive processing. Practically, the product benefits students, lecturers, and institutions by promoting independent learning and supporting digital transformation in education. Overall, the developed multimedia is considered valid, effective, and feasible as an innovative instructional medium for Visual Programming.
Evaluating Learning Management System Adoption Effectiveness for Educators and Students in Batam Higher Education Hendi Sama; Winson Napoleon; Herman
Edu Komputika Journal Vol. 12 No. 2 (2025): Edu Komputika Journal
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/edukom.v12i2.37063

Abstract

The rapid expansion of digital technology and online platforms is reshaping higher education, driving widespread adoption of Learning Management Systems (LMS) and synchronous ICT tools for teaching and learning. As institutions increasingly rely on these technologies, it is essential to understand their effectiveness in supporting educational transformation rather than simply digitizing traditional practices. This study investigates the perceived integration and impact of LMS and ICT tools among students and educators across six universities, using the Technological Pedagogical Content Knowledge (TPACK) framework. A quantitative survey of 579 respondents assessed seven TPACK domains with reliable measurement instruments. Descriptive statistics, t-tests, correlation, and regression analyses were employed to compare perspectives and identify key predictors of successful technology integration. Results reveal that both groups express positive views of LMS and ICT, with all TPACK constructs rated above 3.8. Educators report higher self-assessed competencies, particularly in technological and pedagogical domains, although the differences are statistically significant but practically minor. Strong correlations among TPACK factors and regression modeling highlight Technological Pedagogical Knowledge and Technological Content Knowledge as the most important predictors of effective integration, underscoring the importance of system supported pedagogical design and content delivery rather than simple content digitization. Despite robust infrastructure and readiness, the findings suggest that LMS and ICT are primarily used to support rather than truly transform learning experiences.  The study contributes to educational information systems literature by providing empirical evidence on user perceptions and the need for a system design strategies to achieve more innovative, student centric, technology‑enhanced education in Batam.
Public Sentiment Classification of Danantara in Social Media X Using Support Vector Machine and Random Forest Primandani Arsi; Pungkas Subarkah; Ranggi Praharaningtyas Aji
Edu Komputika Journal Vol. 12 No. 2 (2025): Edu Komputika Journal
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/edukom.v12i2.38254

Abstract

The increasing use of social media as a platform for public discourse provides valuable data for understanding societal responses to national strategic policies. One prominent example is the establishment of Danantara (Daya Anagata Nusantara), a sovereign wealth fund launched by the Indonesian government in February 2025. This study aims to analyze public sentiment toward Danantara using Indonesian-language posts collected from social media platform X and to comparatively evaluate the performance of Support Vector Machine (SVM) and Random Forest (RF) algorithms. A dataset of 1,434 public tweets was collected through web scraping and processed using text preprocessing techniques, including cleaning, tokenization, stopword removal, stemming, and TF-IDF feature extraction. Sentiment labels were generated using an Indonesian RoBERTa model and validated by a linguistic expert. Class imbalance was addressed using the Synthetic Minority Oversampling Technique (SMOTE). Model performance was evaluated using 5-fold stratified cross-validation with accuracy, precision, recall, and F1-score metrics. Experimental results show that Random Forest achieved slightly superior performance, reaching an average accuracy of 91.47%, compared to 91.06% obtained by SVM. Confusion matrix analysis indicates that RF better distinguishes neutral sentiment, while SVM performs competitively in identifying strong sentiment polarity. This study contributes by providing the first empirical comparison of classical machine learning approaches for analyzing public sentiment toward Indonesia’s sovereign wealth fund discourse, offering methodological insights and practical implications for data-driven policy evaluation using social media analytics.
The Impact of Implementing the VAK Model in Practical Learning with Case Method to Improving Students' 21st Century Skills Riska Dami Ristanto; Alfian Ardhiansyah; Ajeng Rahma Sudarni; Humam Nasywa Fawazi; Syarif Romadloni; Damar Hadziq Hidayatilah
Edu Komputika Journal Vol. 12 No. 2 (2025): Edu Komputika Journal
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/edukom.v12i2.39372

Abstract

The advancement of the Industrial Revolution 4.0 era requires university graduates to master 21st-century skills in order to actively contribute and play a role in the workforce. This necessity has become a key factor influencing graduates’ employability. However, challenges arise during the learning process, such as low student participation, limited conceptual understanding, and difficulties in developing critical thinking and problem-solving skills, particularly in practicum activities. A solution to address these issues is to implement a combination of student-centered learning models, namely the VAK (Visualization, Auditory, Kinesthetic) model, in practical learning based on the Case Method. The purpose of this study is to analyze the effect of applying the VAK model in case method–based practical learning on the improvement of students’ 21st-century skills. This research employed a Pre-Experimental Design with a One-Shot Case Study approach. The sample consisted of 94 regular class students of the PTIK Study Program at UNNES in the even semester of 2024/2025, enrolled in the Computer Networks course. The data collection instruments were based on 21st-century skill indicators, including communication, collaboration, critical thinking and problem solving, as well as creativity and innovation. Data analysis was conducted using descriptive and inferential statistics. The findings revealed that the application of the VAK model in case method–based practical learning positively influenced the enhancement of students’ 21st-century skills with a significance value of 0.001. The most significant improvements were observed in Collaboration at 87%, Critical Thinking and Problem Solving at 86.2% with a very strong category. Further research is needed to explore the ideal combination of learning models and approaches to achieve greater improvements in 21st-century skills.
Comparative Evaluation of Statistical and Deep Learning Models for Multi-Product Sales Forecasting Masbahah Masbahah; Kembang Prima Rossari
Edu Komputika Journal Vol. 12 No. 2 (2025): Edu Komputika Journal
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/edukom.v12i2.39797

Abstract

Sales forecasting plays a crucial role in retail decision-making, particularly in multi-product environments with heterogeneous demand characteristics. Unlike most previous studies that focus on finding a single best model globally, this study conducts a standardized comparative evaluation between statistical (TES, SARIMA) and deep learning (LSTM, CNN–LSTM) models using six years of real retail sales data covering eight products with stable seasonal patterns to volatile and nonlinear demand, thus representing the complexity of retail scenarios. To ensure the validity and reliability of the evaluation, data leakage was prevented by splitting the data seasonally (80% training, 20% testing), applying all transformations (scaling, log transform, sequence generation) estimated only on the training data, and using a consistent multi-step autoregressive scheme across all models. Model performance was evaluated using MAPE, MAE, MSE, and RMSE. The adaptive ensemble strategy was implemented by combining SARIMA and CNN–LSTM through optimization of product-specific weights (α ∈ [0,1], interval 0.01) using grid search to minimize MAPE over the testing horizon. The results show that the deep learning model outperforms products with fluctuating and nonlinear demand, while the statistical model remains competitive on products with stable seasonal patterns. Aggregated across products, the adaptive ensemble yielded the lowest average MAPE (0.2184), lower than all individual models, indicating better stability in the face of demand heterogeneity. This study confirms the product-dependent nature of forecasting performance and offers a replicable adaptive framework for multi-product forecasting.  By enhancing data-driven supply chain efficiency and supporting skill development in advanced forecasting methodologies, this research aligns with Sustainable Development Goals (SDG) 8 and SDG 4.
Development of Gamification-Based Mobile Learning to Improve Learning Outcomes for Automotive Measurement Tools Ranu Iskandar; Putri Khoirin Nashiroh; Febrian Arif Budiman
Edu Komputika Journal Vol. 12 No. 2 (2025): Edu Komputika Journal
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/edukom.v12i2.40262

Abstract

Inclusive education, a key priority of SDG 4, requires learning innovations that are accessible and engaging for all learners, including students with disabilities. In Indonesian inclusive schools, limited availability of adaptive learning technology remains a major barrier, particularly for students with mild intellectual disabilities who often struggle with practical competencies such as reading precision measuring tools. This study developed and evaluated a gamification-based mobile learning (m-learning) medium on automotive measurement tools (vernier calipers) tailored to the needs of students with mild intellectual disabilities through accessible design elements (preferred colors, readable fonts, simplified sentences, and adapted assessments). Using a Research and Development (R&D) approach with the ADDIE model, the product was validated by one media expert and one material expert, and its effectiveness was examined through a one-group pretest–posttest design with 35 Grade 10 Automotive Engineering students at SMK Palapa Semarang. Expert validation results indicated that the media was very feasible, with scores of 85% (Display Design), 83% (Technical Quality), 88% (Visual Communication), and 90% (Media Attractiveness), while content feasibility was also very feasible, with 82% (Content Completeness), 87% (Content Accuracy), 84% (Presentation Cohesion), and 86% (Language Use). Student learning outcomes improved substantially, with mean scores increasing from 68.29 (pretest) to 88.86 (posttest), yielding a gain of 20.6 points and reduced score variability (SD from 10.43 to 6.31). These findings suggest that gamification-based m-learning designed using assistive technology principles is feasible and can enhance learning outcomes in inclusive vocational education contexts.
Hybrid CNN-Fuzzy Logic System for Type 2 Diabetes Mellitus Prediction: A Clinical Decision Support Tool with Interpretability Enhancement Anita Lufianti; Kartika Imam Santoso; Meity Mulya Susanti; Rahmawati
Edu Komputika Journal Vol. 12 No. 2 (2025): Edu Komputika Journal
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/edukom.v12i2.40977

Abstract

Type 2 Diabetes Mellitus (T2DM) is a critical global health challenge, with 589 million adults currently diagnosed and projected to reach 853 million by 2050. Early detection is crucial, as it can reduce complication incidence by 30-40%, yet approximately 50% of cases remain undiagnosed. While machine learning approaches demonstrate promise for T2DM risk prediction, current systems face a fundamental accuracy-interpretability paradox: deep learning models achieve high accuracy (88-93%) but lack clinical transparency, while interpretable models sacrifice predictive performance. This study develops and validates a hybrid CNN-Fuzzy Logic system that directly addresses this paradox by combining high predictive accuracy with clinical interpretability. The system employs a Convolutional Neural Network component for non-linear feature abstraction combined with Mamdani Fuzzy Logic incorporating clinically derived weights aligned with ADA 2024 diagnostic criteria. Tested on the Pima Indian Diabetes dataset (n=154 test cases), the hybrid model achieved 92.5% accuracy (95% CI: 88.2-96.1%), 91.2% sensitivity, 93.1% specificity, and AUC-ROC 0.925, statistically superior to standalone CNN (88.9%, p=0.0037) and Fuzzy Logic (88.3%, p=0.0015) approaches. Interpretability scores reached 0.78-0.86, exceeding pure neural network baselines (0.32-0.42) and supporting clinician-understandable risk stratification. The system is operationalized as a web-based Clinical Decision Support System supporting both individual patient assessment and batch population screening. This hybrid architecture directly bridges the accuracy-interpretability paradox that has historically constrained ML adoption in clinical diabetes management.
Customer Review Sentiment Analysis on Google Maps as a Basis for CRM Strategy Optimization: Evidence from Coffeeshop Z Purwokerto Zanuar Rifa'i; Saskia Anjelita; Sunyi Praptaningsih
Edu Komputika Journal Vol. 12 No. 2 (2025): Edu Komputika Journal
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/edukom.v12i2.44940

Abstract

The rapid growth of digital platforms has encouraged customers to share their experiences and evaluations through online reviews, particularly on Google Maps. These reviews provide valuable insights that can support Customer Relationship Management (CRM) strategies when analyzed systematically. This study aims to analyze customer review sentiment toward Coffee Shop Z in Purwokerto using the Multinomial Naïve Bayes algorithm and to utilize the findings as a basis for CRM strategy formulation through the get, keep, and grow framework. A mixed-method approach was employed, combining quantitative sentiment analysis of 242 Google Maps reviews and qualitative interviews with customers for methodological triangulation. The research process included data collection through web scraping, text preprocessing, TF-IDF feature extraction, sentiment labeling, classification modeling, and performance evaluation. The results indicate that negative sentiment dominated customer reviews (57.02%), followed by neutral sentiment (30.16%) and positive sentiment (12.80%). The classification model achieved an accuracy of 63.27%, with stronger performance in identifying negative sentiment than neutral and positive sentiments. The findings reveal that customer dissatisfaction is primarily associated with service quality and waiting time, while positive perceptions are related to food quality, affordability, and cleanliness. Based on these findings, CRM strategies were developed through customer acquisition (get), retention (keep), and loyalty enhancement (grow) initiatives. This study demonstrates that sentiment analysis can serve not only as a customer perception measurement tool but also as a valuable source of evidence-based recommendations for customer relationship management and digital marketing strategy development.

Page 4 of 5 | Total Record : 45

 1   2   3   4   5 

Filter by Year

2024 2026


Reset
Filter By Issues
All Issue Vol. 13 No. 1 (2026): Edu Komputika Journal Vol. 12 No. 2 (2025): Edu Komputika Journal Vol. 12 No. 1 (2025): Edu Komputika Journal Vol. 11 No. 2 (2024): Edu Komputika Journal Vol. 11 No. 1 (2024): Edu Komputika Journal