cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
-
Contact Email
-
Phone
-
Journal Mail Official
semdejafet1908@gmail.com
Editorial Address
-
Location
Kota adm. jakarta timur,
Dki jakarta
INDONESIA
JITK (Jurnal Ilmu Pengetahuan dan Komputer)
Published by STMIK Nusa Mandiri
ISSN : -     EISSN : 25274864     DOI : -
Core Subject : Science,
Computer Science & IT
Kegiatan menonton film merupakan salah satu cara sederhana untuk menghibur diri dari rasa gundah gulana ataupun melepas rasa lelah setelah melakukan aktivitas sehari-hari. Akan tetapi, karena berbagai alasan terkadang seseorang tidak ada waktu untuk menonton film di bioskop. Dengan bantuan media internet, berbagai macam aplikasi nonton film android sangat mudah dicari. Hanya bermodalkan smartphone saja para penonton film dapat streaming berbagai macam jenis film di mana saja dan kapan saja mereka inginkan. Akan tetapi, karena banyaknya pilihan aplikasi nonton film android yang bisa digunakan, terkadang seseorang bingung memilihnya. Untuk itu, diperlukan suatu sistem pendukung keputusan yang dapat digunakan para pengguna sebagai alat bantu pengambilan keputusan untuk memilih dengan berbagai macam kriteria yang ada. Salah satu metode yang digunakan adalah metode Analytical Hierarchy Process (AHP). AHP melakukan perankingan dengan melalui penjumlahan antara vector bobot dengan matrik keputusan dengan tujuan agar hasil yang diberikan lebih baik dalam menentukan alternatif yang akan dipilih. Berdasarkan hasil penelitian yang dilakukan oleh 36 sampel responden didapatkan kriteria konten menjadi prioritas pertama pengguna untuk memilih aplikasi nonton film android dengan nilai bobot sebesar 0,224. Sedangkan Netflix menjadi alternatif dengan prioritas pertama keputusan pengguna dalam memilih aplikasi nonton film android dengan nilai bobot sebesar 0,352.
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 543 Documents
 51   52   53   54   55 
HYBRID DATA MINING METHODS TO SUPPORT MSME SUSTAINABILITY IN RURAL AND URBAN AREAS Krisantus Jumarto Tey Seran; Debora Chrisinta; Yasinta Oktaviana Legu Rema; Hevi Herlina Ullu; Budiman Baso
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 12 No. 1 (2026): JITK Issue August 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v12i1.8271

Abstract

This study aims to analyze the factors influencing the sustainability of MSMEs in North Central Timor Regency by utilizing the K-Mode clustering method and Naive Bayes classification. The data used includes 550 MSMEs in North Central Timor Regency, East Nusa Tenggara Province, classified based on attributes such as location, product price, financial condition, innovation, technology utilization, and sustainability. The K-Mode method was employed to group MSMEs based on categorical similarities after the data was segmented by location attributes, while the Naive Bayes method was applied to classify MSME sustainability following clustering. The results indicate that in rural areas, MSMEs tend to dominate with high product prices and good financial conditions but show low levels of innovation and technology utilization. In contrast, MSMEs in urban areas are generally more innovative and technology-driven despite facing infra-structure challenges. The application of Naive Bayes demonstrated that a data training ratio of 70:30 yielded the best accuracy. Accordingly, the resulting model can be utilized to monitor the sustainability conditions of MSMEs. This study provides insights into sustainability patterns of MSMEs in both rural and urban areas and opens opportunities for further research on external factors affecting sustainability and barriers to technology adoption in rural areas.
REINFORCEMENT LEARNING-BASED DYNAMIC PRICING IN A STOCHASTIC DEMAND–SUPPLY ENVIRONMENT Nur Alamsyah; Budiman; Almira Nurchawilah; Wala Erpurini
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 12 No. 1 (2026): JITK Issue August 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v12i1.8286

Abstract

Dynamic pricing in ride-sharing platforms must balance revenue generation with stable pricing decisions under changing demand and supply. This study aims to develop and evaluate a reinforcement learning-based dynamic pricing policy that maximizes expected revenue while reducing abrupt policy-level price adjustments. A stochastic contextual environment was constructed from 1,000 historical ride records and evaluated using a leakage-safe 70/15/15 train-validation-test split. The agent was trained with Proximal Policy Optimization (PPO) using five discrete price adjustments from -10% to +10%. Expected revenue was combined with a multiplier-based stability penalty, where stability was measured from changes in the price multiplier rather than nominal price variation across heterogeneous rides. Across 30 paired test episodes, the PPO policy achieved a cumulative reward of 103,316.25 +/- 3,243.99 and expected revenue of 103,449.18 +/- 3,241.27, significantly exceeding static pricing (p < 0.001). Relative to rule-based surge pricing, PPO produced statistically indistinguishable cumulative reward (p = 0.808) while reducing multiplier volatility by 24.83%, mean absolute multiplier change by 24.21%, and action switch rate by 10.97% (all p < 0.001). These results indicate that PPO can preserve near-surge revenue while producing smoother dynamic pricing decisions within the simulated environment.
TRANSFORMER-BASED GENERATIVE CHATBOT FOR HIGHER EDUCATION INFORMATION SERVICES WITH HYPERPARAMETER OPTIMIZATION AND MODEL EVALUATION Leni Fitriani; Endang Prayoga Hidayatulloh; Dede Kurniadi; Muhammad Rikza Nashrulloh
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 12 No. 1 (2026): JITK Issue August 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v12i1.8295

Abstract

Generative AI has revolutionized the creation of realistic multimedia content, including chatbots that generate human-like responses. This technology significantly improves higher education by enabling universities to provide fast, accurate, and efficient information services. Generative chatbots can handle multiple users simultaneously and operate 24/7, increasing productivity and accessibility. The model was developed using Machine Learning Lifecycle (MLLC) with deep learning algorithm and Transformer architecture. The dataset used consists of 5,403 question-answer pairs from Institut Teknologi Garut (ITG), which are divided into 5,089 pairs for training and 314 pairs for testing. From 12 hyperparameter configurations, the best combination (maxlen 80, num_layers 2, batch_size 128, embedding_dim 256, fully_connected_dim 256, num_heads 2, positional_encoding_length 512, learning_rate 0.0002, and epoch 100) achieved a BLEU score of 71.03% on the ITG dataset. Evaluation using ROUGE and METEOR also shows consistent performance, indicating good content coverage and semantic similarity. Retraining with another dataset using the same approach resulted in a slightly higher BLEU score of 72.05%, with a different optimal learning rate of 0.00025. The results of this study indicate that Transformer-based generative chatbots can support higher education services and highlight the importance of adjusting hyperparameters based on dataset characteristics. This research also provides opportunities to develop similar models in other universities by adapting datasets and exploring more advanced methods to improve performance in broader educational contexts.
BNI STOCK DIRECTIONAL TREND IDENTIFICATION USING EXPONENTIAL MOVING AVERAGE METHOD angga prastya sianipar; indra kelana; margaretha yohanna
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 12 No. 1 (2026): JITK Issue August 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v12i1.8454

Abstract

High stock volatility demands responsive prediction for risk mitigation. Traditional indicators like Simple Moving Average (SMA) suffer from lagging effects due to equal data weighting, while complex machine learning models often cause overfitting and lack interpretability, constituting a notable research gap. This study addresses these limitations by implementing the Exponential Moving Average (EMA) method with a dual-period configuration (EMA-10 and EMA-20). The EMA is justified by its recursive exponential smoothing that weights recent prices heavily, maximizing sensitivity to abrupt reversals without excessive noise. Utilizing daily closing prices of PT Bank Negara Indonesia (Persero) Tbk (BBNI) from 2020 to 2025, a desktop automated analytical system was developed using Python and PyQt5. Out-of-sample evaluation yields highly stable performance, with Mean Absolute Percentage Errors (MAPE) of 2.45% for EMA-10 and 3.10% for EMA-20, alongside a 92.86% Directional Accuracy. These key findings confirm the precision of the EMA framework in generating Golden Cross and Death Cross signals during trending market phases, although supplementary momentum filters are needed during sideways consolidation. This research provides a robust, objective technical framework that successfully bridges traditional visual analysis and automated investment decision-making.
BLACKBOX TESTING OF THE GIS-BASED INTEGRATED HEALTH POST (POSYANDU) PERFORMANCE MAPPING APPLICATION Riswan Riswan; Nilawati Nilawati; Ahmad Husna Alhadi; Yeni Nurjani; Gustina Gustina; Afrizal Afrizal
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 12 No. 1 (2026): JITK Issue August 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v12i1.7452

Abstract

The Family Planning - Integrated Health Service Post (Posyandu) plays an important role in providing primary health services at the community level in Indonesia, especially in monitoring maternal and child health. However, traditional management and performance evaluation of Posyandu often faces challenges such as limited data accuracy, inefficient reporting, and a lack of integrated monitoring tools. This research aims to develop a Geographic Information System (GIS)-based application to improve the mapping and performance assessment of Posyandu. In contrast to previous GIS health mapping systems, this study makes a unique contribution by integrating automated performance classification based on Posyandu success indicators as well as real-time analytics to support more responsive decision-making. This research uses the Research and Development (R&D) method with the Waterfall approach. The resulting application integrates performance dashboard features, GIS-based location mapping, and data management modules. Blackbox Testing and User Acceptance Test (UAT) are conducted to ensure the functionality and reliability of the system. The functional test results through 10 black box test scenarios show a 100% success rate, while the UAT results achieve an average score of 100% confirming the usability of the system. The app allows cadres to manage data efficiently and provide transparent information to the health office. This study concludes that this GIS-based application significantly improves data management and performance visualization, thereby strengthening the role of Posyandu in public health services.
SIMRS RISK ASSESSMENT USING OCTAVE ALLEGRO METHOD AND ISO/IEC 27001:2022 CONTROL STANDARD Ito Setiawan; Kharisma Putri Sholekha; Retno Waluyo
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 12 No. 1 (2026): JITK Issue August 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v12i1.7670

Abstract

Cilacap Regional General Hospital has been using the SIMRS system since 2006 to the present day. however, its use has encountered several challenges, such as human error, lengthy data synchronisation processes and time-consuming application maintenance. Previous studies have analyzed risk assessments regarding application usage, yet there is a gap in the implementation of risk mitigation.  The aim of the research is to carry out a risk assessment of the SIMRS using the OCTAVE Allegro method and to propose risk mitigation measures in accordance with ISO/IEC 27001. The novelty of this research lies in the combination of methods used to conduct risk assessment and mitigation utilising the ISO/IEC 27001:2022 control standards. Results of research identify the areas of impact, reputation, finance and productivity. The areas of concern are human error, hardware management and software management. Scenarios were developed based on these areas of concern; according to the calculations, the score for human error was 31, for hardware management 24 and for software management 18. Mitigation, as set out in ISO/IEC 27001:2022, focuses on organisational controls in clauses 5.1, 5.15, 5.24, 5.26, 5.28, 5.33, 5.37. People controls in clauses 6.2, 6.3, 6.5, 6.8. Physical controls in clauses 7.5, 7.7, 7.10, 7.12, 7.13. Technological controls in clauses 8.2, 8.5, 8.7, 8.12, 8.13, 8.20, 8.22, 8.32. The scientific contribution lies in the integration of the OCTAVE Allegro and ISO/IEC 27001:2022 approaches to produce a risk assessment process that is not only capable of identifying and prioritising risks, but also directly provides relevant security controls.
BLOCKCHAIN-POWERED ACADEMIC CERTIFICATION SYSTEM USING NEAR PROTOCOL AND NFTS Suwarno Suwarno; Herman; Elvin Valentino
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 12 No. 1 (2026): JITK Issue August 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v12i1.7685

Abstract

Academic certificates play a crucial role in today's competitive job market by helping employers assess candidates' qualifications and suitability for specific roles, thereby enhancing employability. However, traditional methods of certificate verification are fraught with challenges, particularly the widespread issue of forgery. This study proposes a blockchain-based academic certification system using the NEAR Protocol to improve the reliability of certificate issuance and verification. The developed decentralized application (DApp) issues certificates as Non-Fungible Tokens (NFTs), enabling tamper-resistant and easily verifiable credentials. The system was developed using Scrum and evaluated through qualitative thematic analysis of structured interview data. The architecture includes a NEAR-based NFT smart contract, RESTful API, and web frontend. Functional and integration testing on the NEAR testnet confirmed successful certificate minting, storage, and verification with low transaction costs. User evaluation results show that 80.48% of response-level feedback (169/210 responses) was positive across the evaluated themes, with 24 of 30 participants expressing willingness to use the system. Compared with Ethereum and Solana, the proposed DApp system demonstrates better cost efficiency. The novelty of the study lies in integrating Scrum-based development, structured user evaluation, and NFT-based credentials closely into a unified academic certification framework. The proposed approach effectively streamlines certificate management and helps mitigate certificate fraud.
ASSESSING USER SATISFACTION AND ALTERNATIVE DESIGN RECOMMENDATION OF MOBILE BANKING APPLICATIONS Andang Wijanarko; Bagus Mirzana; Aan Erlansari; Yudi Setiawan
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 12 No. 1 (2026): JITK Issue August 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v12i1.7802

Abstract

Advances in information technology in the banking sector have led banking institutions to continuously innovate and improve their services. One such service is the emergence of mobile-based banking applications. Bank Bengkulu has introduced a mobile-based banking application, but it has received poor ratings and negative feedback from its users. This study aims to evaluate user satisfaction with the Bengkulu mobile banking application using the End User Computing Satisfaction (EUCS) method, then improve user satisfaction by redesigning the application using the Double Diamond method, and re-evaluate the results of the application redesign using the System Usability Scale (SUS) method and sentiment analysis using Natural Language Processing (NLP) with the IndoBERT model. The results show that user satisfaction with the Bengkulu mobile banking application was 2.77 when measured using the EUCS method. This measurement falls into the ‘adequate’ category but is very close to the ‘dissatisfied’ category. After the application redesign, user satisfaction improved, as indicated by the evaluation of users using the SUS method with a score of 87.975, which falls into the ‘excellent’ category. The results of qualitative user satisfaction measured using sentiment analysis methods show that there were 75 positive sentiments, 22 neutral sentiments, and only 3 negative comments. The results of the study show that there was a notable improvement in user satisfaction after the Bengkulu Bank application was redesigned.
GENETIC ALGORITHM-BASED HYPERPARAMETER OPTIMIZATION IN DEEP LEARNING FOR HIGH-ACCURACY LOGISTICS EXPENDITURE CLASSIFICATION Marwanto Rahmatuloh; Supriady Supriady; Rukmi Juwita
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 12 No. 1 (2026): JITK Issue August 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v12i1.7840

Abstract

Traditional reactive logistics auditing often fails to detect hidden operational inefficiencies, particularly in transaction data with imbalanced class distributions. This study aims to analyze the effect of Genetic Algorithm (GA)-based hyperparameter optimization on the performance of deep learning models for logistics expenditure efficiency classification and to compare its performance with baseline Artificial Neural Network (ANN) and Convolutional Neural Network (CNN) models. This study used a real-world logistics transaction dataset, with the Cost per Kilogram metric employed as the basis for classification labeling. The research stages included data collection, preprocessing and feature engineering, data balancing using the Synthetic Minority Over-sampling Technique (SMOTE), development of ANN and CNN models, ANN hyperparameter optimization using GA, and comparative evaluation based on accuracy, precision, and recall. The results showed that the baseline ANN and CNN models obtained a recall of 0.00 in detecting inefficient transactions, whereas the GA-optimized ANN achieved an accuracy of 91% and a perfect recall of 1.00. These findings indicate that GA-based hyperparameter optimization improves the model's ability to detect inefficient transactions and reduces diagnostic blind spots in imbalanced logistics financial data. This study contributes a high-accuracy classification approach that can support proactive financial monitoring and automated auditing in the logistics sector.
OVERFITTING MITIGATION AND TRAINING OPTIMIZATION STRATEGIES IN DEEP LEARNING-BASED IMAGE CLASSIFICATION: A SYSTEMATIC REVIEW Budiman Budiman; Arief Setyanto; Andi Sunyoto; Dhani Ariatmanto
JITK (Jurnal Ilmu Pengetahuan dan Teknologi Komputer) Vol. 12 No. 1 (2026): JITK Issue August 2026
Publisher : LPPM Nusa Mandiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33480/jitk.v12i1.7948

Abstract

Overfitting remains a critical challenge in developing deep learning–based image classification models, particularly as modern architectures become increasingly complex and parameter-intensive. Although convolutional and transformer-based models have demonstrated strong predictive performance, their generalization  capability depends strongly on the availability of large, well-annotated, and diverse training datasets. This condition is often difficult to achieve in real-world domains such as medicine, agriculture, and industrial inspection. Previous survey studies have examined various techniques related to image classification, data augmentation, and optimization strategies; however, these studies typically analyse individual approaches in isolation. As a result, opportunities remain to further synthesise the relationships among the underlying causes of overfitting, mitigation strategies, and training parameter configurations within a unified analytical perspective. This study employed the PRISMA framework to conduct a Systematic Literature Review (SLR). A total of 174 primary studies published between 2020 and 2025 were traced to address the gap. The review identifies three major sources of overfitting in image classification tasks: limited labeled data, model architecture complexity, and data and label quality issues. Based on these findings, the study synthesises the corresponding mitigation strategies reported in the literature. The main contribution of this study is not the proposal of a new theoretical taxonomy, but the systematic organisation and synthesis of methodological evidence into an integrated analytical framework. The resulting analytical framework relates the underlying causes of overfitting, mitigation strategies, training optimisation mechanisms, and parameter configuration practices within a unified perspective, thereby facilitating a more integrated interpretation of how these complementary aspects contribute to model generalisation in deep learning–based image classification.

Page 53 of 55 | Total Record : 543

 51   52   53   54   55 

Filter by Year

2016 2026


Reset
Filter By Issues
All Issue Vol. 12 No. 1 (2026): JITK Issue August 2026 Vol. 11 No. 4 (2026): JITK Issue May 2026 Vol. 11 No. 3 (2026): JITK Issue February 2026 Vol. 11 No. 2 (2025): JITK Issue November 2025 Vol. 11 No. 1 (2025): JITK Issue August2025 Vol. 10 No. 4 (2025): JITK Issue May 2025 Vol. 10 No. 3 (2025): JITK Issue February 2025 Vol. 10 No. 2 (2024): JITK Issue November 2024 Vol. 10 No. 1 (2024): JITK Issue August 2024 Vol. 9 No. 2 (2024): JITK Issue February 2024 Vol. 9 No. 1 (2023): JITK Issue August 2023 Vol. 8 No. 2 (2023): JITK Issue February 2023 Vol. 8 No. 1 (2022): JITK Issue August 2022 Vol 8 No 1 (2022): JITK Issue August 2022 Vol 7 No 2 (2022): JITK Issue February 2022 Vol 7 No 1 (2021): JITK Issue August 2021 Vol 6 No 2 (2021): JITK Issue February 2021 Vol 6 No 1 (2020): JITK Issue August 2020 Vol 5 No 2 (2020): JITK Issue February 2020 Vol 5 No 1 (2019): JITK Issue August 2019 Vol 4 No 2 (2019): JITK Issue February 2019 Vol 4 No 1 (2018): JITK Issue August 2018 Vol 3 No 2 (2018): JITK Issue February 2018 Vol 3 No 1 (2017): JITK Issue August 2017 Vol 2 No 2 (2017): JITK Issue February 2017 Vol 2 No 1 (2016): JITK Issue August 2016 More Issue