cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Ari Zulsafar
Contact Email
zulsapar@telkomuniversity.ac.id
Phone
+6285280983983
Journal Mail Official
journals@telkomuniversity.ac.id
Editorial Address
Jl. Telekomunikasi No.1, Sukapura, Kec. Dayeuhkolot, Kabupaten Bandung, Jawa Barat
Location
Kota bandung,
Jawa barat
INDONESIA
IJoICT (International Journal on Information and Communication Technology)
Published by Universitas Telkom
ISSN : -     EISSN : 23565462     DOI : https://doi.org/10.21108/ijoict
Core Subject : Science, Social,
Computer Science & IT Environmental Science
nternational Journal of Information Communication Technology (IJoICT) is a peer-reviewed Journal. This journal includes novel ideas on ICT, state of the art technique implementations, and study cases on developing countries. This journal fully acknowledges the articles that emphasize a balanced coverage between theory and practice. Subject areas that is suitable for publication to the following fields: Computer Networking and Communication Graphics & Multimedia Theoretical CS & Statistic Embeded System Software Engineering Information System Security & Cryptography Data Science Parallel and Distributed Systems Database Systems Intelligence System
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 17 Documents
 1   2 
Enhancing Digital Forensics with Cyber Kill Chain and 5W1H: A Case Study on Phishing Attacks Erika Ramadhani; Toto Raharjo
IJoICT (International Journal on Information and Communication Technology) Vol. 11 No. 1 (2025): Vol. 11 No. 1 Jun 2025
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/ijoict.v11i1.8966

Abstract

This research has combined the Cyber Kill Chain (CKC) model and the 5W1 H for detection and control of cybercrime such as phishing for the automation of digital forensic investigation. The most vital challenge in digital forensics is its evidence handling complexity, the lack of a standard because of diversified kinds of tools, and the non-availability of automated tools that systematically present information. Therefore, it provides a web-based framework to automate the investigation by referring to the attack stages of the CKC and identifies the contextual allegories of the incident like who, what, when, where, why, and how through the rule of 5W1H. It includes the problem identification method, collecting and classifying the digital artifacts according to CKC stages, in-depth analysis with the 5W1H framework, and visualization of investigation results for further understanding. A case study of a phishing attack on the Kredivo application was used to evaluate the effectiveness of this approach, where the CKC stages from reconnaissance to actions on objectives were implemented to analyze artifacts such as activity logs and phishing data. The results show that the integration of CKC and 5W1H improves analysis accuracy, generates comprehensive visualizations of artifacts, and strengthens response to attacks. It is expected that this finding would mean a highly significant change in the productivity of forensic investigations by making it easier for analysts and preparing proper documentation education for the court.
Pose Classification in Archery Sports Based on YoloV8 Using SVM and Random Forest Methods Yuridikta Adha Muslim; Bedy Purnama; Bayu Erfianto
IJoICT (International Journal on Information and Communication Technology) Vol. 11 No. 1 (2025): Vol. 11 No. 1 Jun 2025
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/ijoict.v11i1.8996

Abstract

This research creates a YOLOv8-based pose classification system that can analyze and classify the movements of archery athletes. The system is combined with SVM and RF methods, and utilizes YoloV8 pose detection and machine learning techniques to provide more accurate classification. Video data collection, system design and implementation, and analysis of implementation results are some of the stages passed during system development. The process includes joint feature extraction using YOLOv8 and classification for Recurve and Barebow categories using SVM and RF. The test results show the difference in performance between the two classification methods. For the Recurve category, SVM had 90% accuracy for testing, while RF had 87% accuracy for testing. For the Barebow category, SVM had 76% accuracy for testing, while RF had 75% accuracy for testing. In terms of generalization, the two methods differed, with SVM showing better stability between testing and training performance. The results show that SVM is superior when testing when compared to RF which makes an anomaly with previous studies
Benchmarking Mobile Apps Security in Universities: An OWASP Mobile Top 10 Framework Perspective Fajar Maulana Kadir; Muhamad Irsan; Aji Gautama Putrada
IJoICT (International Journal on Information and Communication Technology) Vol. 11 No. 1 (2025): Vol. 11 No. 1 Jun 2025
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/ijoict.v11i1.9094

Abstract

Leading Indonesian universities such as Telkom University (Tel-U), Institut Teknologi Bandung(ITB), Universitas Indonesia (UI), and Universitas Gadjah Mada (UGM) have developed mobilebasedacademic information systems that improve the accessibility of campus services, wheresensitive information such as personal data, access credentials, and educational information arestored and managed through the mobile application. The current gap is the lack of understanding ofthe specific vulnerability profile of campus mobile applications and how these vulnerabilities canaffect the data security of educational institutions. This study conducts a comparative analysis ofvulnerabilities in campus mobile applications using the OWASP Mobile Top 10 framework as itstesting standard. In its implementation, this study uses three mobile application security testingtools: AndroBugs, Mobile Security Framework (MobSF), and QARK (Quick Android Review Kit).These three tools were chosen because of their ability to detect various types of vulnerabilitiescovered in the OWASP Mobile Top 10. By comparing vulnerability analysis results on differentcampus mobile applications, this study aims to identify common vulnerability patterns and providerecommendations for improvements following the OWASP Mobile Top 10 security standards. Thetest results show that MySIX ITB and WeAreUI have the most vulnerabilities compared to the otherthree campuses, with 24 vulnerabilities from three different tools. However, if we look at theconsensus between the three tools, MySIX ITB is the most vulnerable application, withvulnerabilities in five categories: M3, M5, M6, M8, and M9. In addition to using three differenttools to strengthen the vulnerability detection rate, we also found some new knowledge. The first isthat all three tools have the same agreement for detecting M2, M6, and M8, which shows the highreliability of the three tools for the categories mentioned. The second is the knowledge that QARKmakes the most different decisions from the other two tools. The test results show that QARK makesdifferent decisions eight times. We also learned that for the four campus mobile apps, developersshould pay more attention to two categories detected by each tool, namely M6 and M8, or InadequatePrivacy Controls and Security Misconfiguration, respectively. Finally, there is knowledge that thestrength of the four mobile apps is resistance to M2; in other words, each campus has used thirdpartylibraries well.
Ant Colony Optimization for Low-Rank Factorization with DNN on People Counting IoT using Environmental Sensors Ganendra Zefanya Patty Patty; Muhammad Faris Fathoni, S.T., M.T., Ph.D. Fathoni; Aji Gautama Putrada Putrada
IJoICT (International Journal on Information and Communication Technology) Vol. 11 No. 1 (2025): Vol. 11 No. 1 Jun 2025
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/ijoict.v11i1.9102

Abstract

People counting Internet of Things (IoT), which plays a role in counting people indoors based on sensor values, is a vital part of smart buildings because it affects other IoT systems that regulate devices like lighting and air conditioning (AC), impacting efficiency. However, a lightweight solution is needed to perform people counting without threatening personal privacy. This study aims to develop an edge computing-based people counting system using environmental sensors and a Deep Neural Network (DNN) model optimized using the LRF technique. The system is designed to operate in real-time on edge devices with low latency and efficient resource consumption. In general, the system's work process is divided into three main stages, namely (1) data acquisition and pre-processing, (2) model development and optimization, and (3) overall system performance evaluation. The system runs automatically on edge devices and follows a cyclic workflow to detect the number of people continuously. This study also uses ant colony optimization (ACO) for hyperparameter tuning and obtains optimum hyperparameters. Experimental results support the claim that LRF significantly reduces model size while maintaining high prediction accuracy. ACO on hyperparameter tuning obtains the optimum hyperparameters: the number of neurons as many as 128 units, Adam learning rate of 0.005, and batch size of 8. Then DNN + ACO is proven to perform better than DNN without ACO and the state-of-the-art random forest model with accuracy, precision, recall, and F1-score of 0.98, 0.99, 0.94, and 0.97. This is while overcoming the imbalance problem in the dataset with recall for counts 0, 1, 2, and 3, of 1.00, 1.00, 1.00, and 0.78, respectively. Finally, we found that the optimum rank on LRF to reduce the number of parameters in DNN is 32, where at that rank the model size is reduced from 28.6 KB to 26.6 KB without significant accuracy loss.
StuntCare: Digital Innovation for Early Warning of Stunting-Risk Families in Sigi Regency Maulidyani Abu; Moh.Al-fath Salsabilah; Juni Wijayanti Puspita; Resnawati; Abdul Mahatir Najar; Rina Ratianingsih; Agus Indra Jaya; Abunawas Tjaija
IJoICT (International Journal on Information and Communication Technology) Vol. 11 No. 1 (2025): Vol. 11 No. 1 Jun 2025
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/ijoict.v11i1.9111

Abstract

The Prevalence of Stunting in Sigi Regency remains notably high at 36.8%, significantly above the national target. Stunting is frequently caused by recurrent infections, poor sanitation, and chronic nutritional deficiencies. Since stunting is a condition of chronic malnutrition that impairs a child's physical and cognitive development, an early warning system is essential for prevention. This study proposes the development of a web-based application to predict the risk of stunting in vulnerable families. Families are the primary focus as they serve as the first environment where children grow and develop. If risk factors are present within a family, the likelihood of stunting increases. Therefore, early detection is crucial for mapping family health conditions. By predicting stunting risks, families can take preventive measures before the condition severely impacts the child. This early warning system serves as a critical alarm, encouraging families to be more vigilant in maintaining the health of all household members. The stunting prediction system is developed as a web-based application, utilizing 11 variables for early stunting detection and employing the K-Nearest Neighbor (K-NN) method. The model's accuracy is evaluated using a Confusion Matrix, achieving an accuracy rate of 99.991%. Keywords: Early Warning System, Stunting, Classification, K-Nearest Neighbor, Confusion Matrix
Association Analysis Between Public Sentiment and Grab Stock Performance Using SVM and Lambda Test Dita Pramesti; Hanif Fakhrurroja; Rahma Karina M.
IJoICT (International Journal on Information and Communication Technology) Vol. 11 No. 1 (2025): Vol. 11 No. 1 Jun 2025
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/ijoict.v11i1.9152

Abstract

During a period of strong economic performance in Indonesia—marked by a 5.4% growth in the second quarter of 2022—concerns about a potential downturn in the fourth quarter began to surface, as indicated by increased stock market volatility, including fluctuations in Grab’s share prices. This study aims to classify public sentiment toward Grab based on comments from the social media platform Twitter, and to analyze its relationship with the direction of the company’s stock price movement. Sentiment classification was conducted using the Support Vector Machine (SVM) algorithm through a series of steps including data preprocessing, TF-IDF weighting, imbalance data handling, and model performance evaluation. The dataset was split into 70% training data and 30% testing data. The SVM model achieved an accuracy of 87%, with a precision of 90%, recall of 91%, and F1-score of 91%. Public sentiment for each period was then aggregated using the Net Sentiment Score (NSS), which was subsequently categorized into positive or negative sentiment. These sentiment categories were analyzed in relation to stock price movements using the Goodman-Kruskal Lambda test. The result of ????(stock∣sentiment)=0.053 indicates that knowing public sentiment reduces prediction error by only 5.3%, while ????(sentimen|saham)=0.000 shows no predictive value in the opposite direction. This study contributes a novel approach by integrating machine learning-based sentiment classification with a categorical association test, specifically applied to a regional technology company in Southeast Asia, which remains underexplored in existing literature.
Analysis of Public Sentiment Regarding the 2024 Jakarta Election on Platform X Using Deep Learning Moch. Rizki Khaerul Muhaemin; Fitriyani; Lazuardy Syahrul Darfiansa
IJoICT (International Journal on Information and Communication Technology) Vol. 11 No. 1 (2025): Vol. 11 No. 1 Jun 2025
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/ijoict.v11i1.9176

Abstract

The 2024 Jakarta Regional Head Election (Pilkada Jakarta) is a critical issue that requires an in-depth understanding of public sentiment. This platform generates complex, unstructured text with informal language and ambiguity, posing challenges alongside the lack of local context-specific datasets and inaccuracies in traditional sentiment analysis models. Analyzing sentiment for the Pilkada is crucial for evaluating public response to policies, aiding political strategy, and improving governance. Current systems struggle with complex data and class imbalance (dominant neutral sentiment), leading to underrepresented information. This study addresses these issues by constructing a sentiment analysis system using four deep learning models: IndoBERT, LSTM, CNN, and GRU. The procedure encompassed data acquisition from X, preprocessing, model training, and assessment based on accuracy, precision, recall, and F1-score. The CNN model achieved the highest accuracy of 83.37%, followed by LSTM at 82.61%, GRU at 82.30%, and IndoBERT at 80.77%. All models achieved the accuracy benchmark of a minimum of 80%, however the neutral class continues to pose a challenge. Research contributions include a deep learning-based sentiment classification system that can be implemented in local political opinion analysis, as well as recommendations for using hybrid models like IndoBERT + CNN for further research.
A Comparative Study on Handling Imbalanced Data in Indonesian Hate Speech Detection Using FastText and BiLSTM Akmal Muhamad Faza; Yuliant Sibaroni; Sri Suryani Prasetiyowati
IJoICT (International Journal on Information and Communication Technology) Vol. 11 No. 2 (2025): Vol. 11 No. 2 Dec 2025
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/ijoict.v11i2.9513

Abstract

Online hate speech has become a serious threat to social harmony in Indonesia, with cases increasing significantly in recent years. This study develops and evaluates a system for detecting Indonesian hate speech using a Bidirectional Long Short-Term Memory (BiLSTM) deep learning model, complemented by FastText word embeddings. To address the common issue of data imbalance in hate speech datasets, this study implements and compares three oversampling techniques: Random Oversampler, Synthetic Minority Oversampling Technique (SMOTE), and Adaptive Synthetic Sampling (ADASYN). The research utilizes the Indonesian Hate Speech Superset, a dataset comprising 14,306 comments. The model's performance is evaluated using Stratified K-fold Cross-Validation, with metrics including Accuracy, Precision, Recall, and F1-score. Results, visualized using a Confusion Matrix to demonstrate that applying oversampling techniques enhances model performance, particularly by improving the Recall and F1-score metrics. These findings contribute to the development of hate speech classification systems that are fairer, more adaptive, and better suited to the unique characteristics of the Indonesian social media landscape.
Geospatial Sentiment Analysis of Negative Comments on the 2024 Election Using the Robustly Optimized BERT Approach (RoBERTa)Geospatial Sentiment Analysis of Negative Comments on the 2024 Election Using the Robustly Optimized BERT Approach (RoBERTa)Geospat Haidar ali; Yuliant Sibaroni
IJoICT (International Journal on Information and Communication Technology) Vol. 11 No. 2 (2025): Vol. 11 No. 2 Dec 2025
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/ijoict.v11i2.9575

Abstract

This study develops a geospatial sentiment analysis system to detect and map hate speech related to the 2024 Election using the Robustly Optimized BERT Approach (RoBERTa). The dataset consists of 11,903 social media comments that have undergone comprehensive preprocessing, including text normalization, stopword removal, and stemming. The RoBERTa model was implemented using 10-fold cross-validation for multi-class classification (HS_Weak, HS_Strong, Not_Abusive) and achieved an average accuracy of 91.54% (±1.08%), with a final model accuracy of 94.29%. Geospatial analysis using geocoding and Folium visualization revealed that 75% of the data originated from Indonesia, with the highest concentration in the Jakarta area. The distribution of hate speech showed consistent patterns between Indonesia (45.6% hate speech) and outside Indonesia (44.3% hate speech), with the HS_Strong category dominating at 96.4%. Heatmap analysis identified hate speech hotspots on the island of Java and a global distribution across various continents. The findings confirm the effectiveness of RoBERTa for sentiment analysis in the Indonesian language and provide valuable insights into the geographic patterns of hate speech in the context of digital politics, which can be used to develop mitigation strategies and real-time monitoring systems.
Prediction and Classification of Vehicle Traffic Congestion in Bandung City Using the Random Forest and K-Nearest Neighbour Algorithm Muhammad Alauddin Angka Kurniawan; Sri Suryani Prasetiyowati; Yuliant Sibaroni
IJoICT (International Journal on Information and Communication Technology) Vol. 11 No. 2 (2025): Vol. 11 No. 2 Dec 2025
Publisher : School of Computing, Telkom University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21108/ijoict.v11i2.9602

Abstract

Traffic congestion remains one of the problems that continue to arise, especially in urban areas, oneof which is Bandung City, when the causes of the problem are not managed properly. Continuousmanagement of the causes of congestion problems will result in a controlled traffic system for theforeseeable future. This condition can be achieved if there is a congestion classification predictionsystem available. A reliable prediction and classification system can support the government informulating data-based traffic management strategies. The Random Forest and K-NearestNeighbour machine learning classification methods are strengthened with time-based featureexpansion to capture traffic behavior in various time frames, so that the objectives can be achieved.The dataset obtained from Area Traffic Control System Bandung includes traffic flow recorded at15-minute intervals at several intersections. Additional features such as red light duration, roadwidth, and spatial proximity to residential and commercial areas are included to improve modelperformance. The results show that the Random Forest classifier with time-based feature expansionoutperforms K-Nearest Neighbors, achieving the highest performance of 96%. These results showthe potential contribution in short-term traffic prediction and its effectiveness in supporting urbantraffic planning and congestion mitigation efforts in Bandung.

Page 1 of 2 | Total Record : 17

 1   2 

Filter by Year

2025 2025


Reset
Filter By Issues
All Issue Vol. 11 No. 2 (2025): Vol. 11 No. 2 Dec 2025 Vol. 11 No. 1 (2025): Vol. 11 No. 1 Jun 2025