Mobile and Forensics

Development of Web Application for Certificate Automation using Cloudflare DNS API and ZeroSSL

2025-12-10T02:42:05+00:00

Existing SSL/TLS automation tools were largely server-centric and required host-level configuration. These limitations increased complexity in distributed and multi-domain environments. This study proposed a centralized web-based SSL certificate automation system. The system integrated the Cloudflare DNS API and ZeroSSL REST API for certificate lifecycle management. The research adopted the Agile Scrum methodology during system development. An asynchronous queue-based architecture was implemented to support concurrent certificate issuance. The architecture reduced API rate-limit constraints. Automated Domain Validation (DV) was successfully performed through DNS integration. The system centralized certificate storage and monitoring. The interface simplified administrator operations. Configuration errors were reduced during certificate deployment. Operational efficiency was improved for distributed infrastructures. A usability evaluation was conducted with 25 technical practitioners. The evaluation produced a System Usability Scale mean score of 83.5 with a standard deviation of 6.7. The findings indicated excellent user acceptance and system usability. In conclusion, the proposed system effectively automated SSL certificate management through a centralized and usability-oriented approach. The system also minimized operational overhead and dependency on host-level configuration.

Machine Learning-Based Lifestyle Analysis for Health Risk Detection in Coffee Drinkers

2026-04-23T00:19:47+00:00

This study aims to develop a machine learning-based intelligent system to detect health risks in coffee drinkers through a lifestyle analysis approach. The background of this study is based on the increasing consumption of coffee as part of a modern lifestyle, which has the potential to cause various health risks if not balanced with a healthy lifestyle. The dataset was collected through a survey covering several important variables, such as coffee consumption frequency, sugar intake, sleep duration, physical activity level, body mass index (BMI), and blood pressure.

The research stages included data preprocessing, normalization, and classification using three algorithms: Random Forest and Gradient Boosting. Model performance was evaluated using accuracy, precision, recall, and F1-score metrics to ensure the system’s reliability. Test results showed that the Random Forest algorithm performed best with an accuracy of 0.91, followed by Gradient Boosting with an accuracy of 0.90.

Further analysis revealed that the variables most influential on health risks are coffee consumption frequency, sleep duration, and sugar intake. The developed system proved effective in detecting health risks early and has the potential to serve as a data-driven educational tool to raise public awareness of the importance of a healthy lifestyle.

Combining RFM-Based Customer Segmentation and K-Means Clustering to Optimize E-Commerce Marketing Strategies

2026-02-22T07:16:28+00:00

This research examines the shift of consumers from offline shopping to online shopping through e-commerce platforms, driven by convenience, product variety, discounts, and rapid technological advancements such as the internet, smartphones, big data, and machine learning. The study utilizes a transactional dataset containing attributes such as InvoiceNo, StockCode, Description, Quantity, InvoiceDate, UnitPrice, CustomerID, and Country to perform customer segmentation. The segmentation process is conducted using the K-means clustering algorithm combined with the RFM (Recency, Frequency, Monetary) model to develop comprehensive customer profiles. This approach aims to identify distinct customer groups and support the creation of more targeted and personalized marketing strategies. The quality of clustering results is evaluated using the Silhouette Score to ensure optimal grouping performance. The entire data processing, analysis, and modeling are implemented using Python as the primary tool. The research methodology includes literature review, data collection, data preprocessing, clustering model development, customer profile analysis, and evaluation and validation of results. The findings are expected to provide valuable insights into customer behavior, enabling e-commerce businesses to enhance marketing effectiveness, improve customer retention, and attract new customers. Ultimately, this study contributes to the development of data-driven marketing strategies that improve customer satisfaction and business performance in the digital commerce environment.

A Hybrid Machine Learning and Digital Forensics Framework for Detecting Coordinated Suspicious Accounts on X

2026-05-29T02:39:19+00:00

Coordinated Inauthentic Behavior (CIB) on X threatens public discourse, yet single-method approaches fail to detect implicit coordination. This study proposes a conditional hybrid framework spanning four forensic stages: (1) automated classification using multilingual BERT text embeddings concatenated with 21 numerical features, (2) sockpuppet candidate generation via Sentence-BERT stylometric fingerprinting and cosine similarity, (3) graph-based coordination analysis using the Louvain algorithm, and (4) manual forensic confirmation with OSINT investigation. Trained on the Cresci-2017 benchmark which exhibits substantial temporal and linguistic domain mismatch from our 2025 Indonesian target domain, the BERT model achieved an accuracy, precision, recall, and F1-score of 1.00 on its held-out test partition prior to deployment. The framework was applied to 1,309 Indonesian Military Law (UU TNI) tweets from 720 accounts, which resulted in the identification of 114 suspicious accounts (15.8%) and the extraction of 28 candidate accounts from 6 groups. Manual validation by the authors through side-by-side timeline inspection confirmed one pair of accounts with synchronized and verbatim identical content. The absence of direct network interactions despite high stylometric similarity is interpreted as consistent with implicit CIB evasion strategies, but attribution to organized groups cannot be definitively established on the basis of platform data alone.

Artefact-to-Legal Element Mapping in iOS-Based Web Gambling Cases: A Cross-Source Correlation Framework under KUHP and ITE Law

2026-03-05T09:33:38+00:00

The widespread use of web-based online gambling through mobile browsers and instant messaging applications has made digital traces increasingly difficult to interpret as legally admissible evidence. Previous forensic research primarily concentrated on artefact recovery from dedicated gambling applications, leaving limited attention to systematic approaches for associating web-based traces with legal elements. This study examined the identification, classification, and legal mapping of web-based gambling artefacts using the proposed Artefact-to-Legal Element Mapping Framework (ALMF). A real-case evidence set acquired from an Apple iPhone XR was analysed with Magnet AXIOM Process v9.4.0.44917, generating Quick Image and Decrypted evidence for extraction and correlation. The analysis identified four major artefact categories, including web browsing traces, credential and persistence artefacts, WhatsApp communication records, and visual media files. Browser artefacts reflected intentional discovery and access to gambling websites, whereas credential artefacts indicated recurring or sustained access. WhatsApp conversations provided the strongest evidence of promotional activities through shared links, invitations, and dissemination patterns. Visual artefacts further strengthened the gambling context by revealing website interfaces and financial indicators. Cross-source correlation differentiated player activities from promoter or distributor behaviour. The mapped evidence supported KUHP Article 303 and UU ITE Article 27(2). The findings demonstrated that contextual interpretation across multiple artefact sources produced stronger evidential value than analysing individual artefacts separately.