cover
Contact Name
Tommy
Contact Email
lpkdgeneration2022@gmail.com
Phone
+6285695565558
Journal Mail Official
tommy@admi.or.id
Editorial Address
Perumahan Bumi Dirgantara Permai Blok CL NO 5, Jl. Durian, Jati Asih, Bekasi, Provinsi Jawa Barat, 17421
Location
Kab. bekasi,
Jawa barat
INDONESIA
International Journal Science and Technology (IJST)
ISSN : 28287223     EISSN : 28287045     DOI : https://doi.org/10.56127/ijst.v1i2
International Journal Science and Technology (IJST) is a scientific journal that presents original articles about research knowledge and information or the latest research and development applications in the field of technology. The scope of the IJST Journal covers the fields of Informatics, Mechanical Engineering, Electrical Engineering, Information Systems and Industrial Engineering. This journal is a means of publication and a place to share research and development work in the field of technology.
Articles 8 Documents
Search results for , issue "Vol. 3 No. 3 (2024): November: International Journal Science and Technology" : 8 Documents clear
DEEP LEARNING METHODS COMPARISON ON IMAGES OF TOMATO AND CUCUMBER LEAF IDENTIFICATION KIRNAP, Ahmet; BİNGÖL, Mehmet Safa; ŞAHİN, Fikri
International Journal Science and Technology Vol. 3 No. 3 (2024): November: International Journal Science and Technology
Publisher : Asosiasi Dosen Muda Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56127/ijst.v3i3.1683

Abstract

Deep learning and image processing applications have become widespread, thanks to hardware developments and increased processing power. The use of technology in agriculture is increasing rapidly with the development of technology. One of the recent applications of technology in agriculture is image processing applications using deep learning. Image processing is aimed at sustainable agriculture. Deep learning is used in applications such as disease detection, agricultural spraying, maturity granding, irrigation, fertilization. In this study, deep learning models AlexNet and SqueezeNet are used to classify tomato and cucumber leaf images. 30 tomato leaves and 30 cucumber leaves are photographed to create the dataset used in the study. Afterwards, the images obtained are increased with data augmentation methods and a data set is created. The dataset consists of 2 classes and a total of 300 images. The data set is used 70% for training and 30% for validation. The results obtained from AlexNet and SqueezeNet deep learning models are given comparatively.
ENHANCING EFFICIENCY AND TRANSPARENCY IN COFFEE SUPPLY CHAIN THROUGH BLOCKCHAIN-INTEGRATED TRACEABILITY PLATFORM Jagad Raya Ramadhan; Donny Avianto
International Journal Science and Technology Vol. 3 No. 3 (2024): November: International Journal Science and Technology
Publisher : Asosiasi Dosen Muda Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56127/ijst.v3i3.1686

Abstract

Coffee is a global commodity that plays a significant role in the economy of many countries, including Indonesia. As the world's fourth-largest coffee producer, Indonesia has a vast potential to increase its coffee exports. This economic impact is not only a source of foreign exchange but also a significant source of income for smallholder farmers. However, recent inefficiencies have led to declining exports and quality control issues. This issue is exacerbated by the lack of transparency and traceability in the coffee supply chain, which makes it difficult for stakeholders to monitor the movement of coffee beans from farm to market. Thus, this research aims to address these problems by developing a blockchain-integrated traceability platform enhanced with IoT technology. The platform connects all stakeholders in the coffee supply chain, including farmers, processors, distributors, sellers, and consumers, ensuring real-time monitoring and data transparency throughout the coffee supply chain. This benefitted not only the involved stakeholders but also the end consumers. The system's provided QR code allows consumers to access information about the coffee's origin, quality, and processing details, increasing customer awareness and trust in the product.
DESIGN AND IMPLEMENTATION OF A SUPERMARKET MANAGEMENT SYSTEM USING UML FOR STREAMLINED INVENTORY, SALES, AND CUSTOMER MANAGEMENT Erni Wigati
International Journal Science and Technology Vol. 3 No. 3 (2024): November: International Journal Science and Technology
Publisher : Asosiasi Dosen Muda Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56127/ijst.v3i3.1712

Abstract

The growing demands on supermarket management have necessitated the adoption of robust systems to streamline operations such as inventory control, sales tracking, and customer management. This study introduces a UML-based design for a supermarket management system that addresses the core needs of store administration through structured modules and defined workflows. The system encompasses store registration, inventory management, sales tracking, and customer engagement, further facilitating seamless product browsing, checkout, and post-sale support. Through structured modeling, the proposed system simplifies process flows, thereby enhancing the operational efficiency and user experience in a supermarket environment. This paper details the UML diagrams used, outlines the functional processes, and discusses the potential benefits of the system for modern retail management.
COMPARISON OF PRE-TRAINED BERT-BASED TRANSFORMER MODELS FOR REGIONAL LANGUAGE TEXT SENTIMENT ANALYSIS IN INDONESIA Taufiq Dwi Purnomo; Joko Sutopo
International Journal Science and Technology Vol. 3 No. 3 (2024): November: International Journal Science and Technology
Publisher : Asosiasi Dosen Muda Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56127/ijst.v3i3.1739

Abstract

This study compared the performance of eight pre-trained BERT-based models for sentiment analysis across ten regional languages in Indonesia. The objective was to identify the most effective model for analyzing sentiment in low-resource Indonesian languages, given the increasing need for automated sentiment analysis tools. The study utilized the NusaX dataset and evaluated the performance of IndoBERT (IndoNLU), IndoBERT (IndoLEM), Multilingual BERT, and NusaBERT, each in both base and large variants. Model performance was assessed using the F1-score metric. The results indicated that models pre-trained on Indonesian data, specifically IndoBERT (IndoNLU) and NusaBERT, generally outperformed the multilingual BERT and IndoBERT (IndoLEM) models. IndoBERT-large (IndoNLU) achieved the highest overall F1-score of 0.9353. Performance varied across the different regional languages. Javanese, Minangkabau, and Banjar consistently showed high F1 scores, while Batak Toba proved more challenging for all models. Notably, NusaBERT-base underperformed compared to IndoBERT-base (IndoNLU) across all languages, despite being retrained on Indonesian regional languages. This research provides valuable insights into the suitability of different pre-trained BERT models for sentiment analysis in Indonesian regional languages.
ANALYSIS OF PARENT-CHILD INTERNET ADDICTION TEST IN SDIT AL IMAN BINTARA Tissa Maharani
International Journal Science and Technology Vol. 3 No. 3 (2024): November: International Journal Science and Technology
Publisher : Asosiasi Dosen Muda Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56127/ijst.v3i3.1770

Abstract

The development of information and communication technology and the implementation of distance learning during the COVID-19 pandemic have made everyone use gadgets, including children and toddlers. In addition to having a positive impact, of course, there are many negative impacts, one of which is gadget addiction. Gadget addiction can be detected using the PARENT-CHILD INTERNET ADDICTION TEST (PCIAT) developed by Dr. Kimberly Young based on Internet Addiction Test (IAT). The purpose of this research is to detect and analyze the use of gadgets and the internet using PCIAT in SDIT Al Iman Bintara students through respondent, namely their parents, and increase the awareness of parents about the dangers of gadget and internet addiction. The results of this research are, from 516 students, as many as 472 student guardians filled out the PCIAT questionnaire. A total of 288 respondent children (67.4%) showed NO SYMPTOMS of gadget addiction, 124 respondent children (29%) showed MILD symptoms, and 15 respondent children (3.5%) showed moderate symptoms. Cooperation between parents and the school is needed in regulating the use of children's gadgets, and consistency in the implementation of the rules.
Efficient TinyML Architectures for On-Device Small Language Models: Privacy-Preserving Inference at the Edge Mangesh Pujari; Anshul Goel; Anil Kumar Pakina
International Journal Science and Technology Vol. 3 No. 3 (2024): November: International Journal Science and Technology
Publisher : Asosiasi Dosen Muda Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56127/ijst.v3i3.1958

Abstract

Deploying small language models (SLMs) on ultra-low-power edge devices requires careful optimization to meet strict memory, latency, and energy constraints while preserving privacy. This paper presents a systematic approach to adapting SLMs for Tiny ML, focusing on model compression, hardware-aware quantization, and lightweight privacy mechanisms. We introduce a sparse ternary quantization technique that reduces model size by 5.8× with minimal accuracy loss and an efficient federated fine-tuning method for edge deployment. To address privacy concerns, we implement on-device differential noise injection during text preprocessing, adding negligible computational overhead. Evaluations on constrained devices (Cortex-M7 and ESP32) show our optimized models achieve 92% of the accuracy of full-precision baselines while operating within 256KB RAM and reducing inference latency by 4.3×. The proposed techniques enable new applications for SLMs in always-on edge scenarios where both efficiency and data protection are critical.
Neuro- Symbolic Compliance Architectures: Real-Time Detection of Evolving Financial Crimes Using Hybrid AI Anil Kumar Pakina; Mangesh Pujari
International Journal Science and Technology Vol. 3 No. 3 (2024): November: International Journal Science and Technology
Publisher : Asosiasi Dosen Muda Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56127/ijst.v4i1.1961

Abstract

This paper proposes NeuroSym-AML, a new neuro-symbolic AI framework explicitly designed for the real-time detection of evolving financial crimes with a special focus on cross-border transactions. By combining Graph Neural Networks (GNNs) with interpretable rule-based reasoning, our system dynamically adapts to emerging money laundering patterns while ensuring strict compliance with FATF/OFAC regulations. In contrast to static rule-based systems, NeuroSym-AML shows better performance-an 83.6% detection accuracy to identify financial criminals, which demonstrated a 31% higher uplift compared with conventional systems-produced by utilizing datasets from 14 million SWIFT transactions. Furthermore, it is continuously learning new criminal typologies, providing decision trails that are available to regulatory audit in real-time. Key innovations include: (1) the continuous self-updating of detection heuristics, (2) automatic natural language processing of the latest regulatory updates, and (3) adversarial robustness against evasion techniques. This hybrid architecture bridges the scalability of machine learning with interpretability of symbolic AI, which can address crucial gaps for financial crime prevention, therefore delivering a solution for satisfying both adaptive fraud detection and transparency in decision-making in high-stakes financial environments.
Ensuring Responsible AI: The Role of Supervised Fine-Tuning (SFT) in Upholding Integrity and Privacy Regulations Tejaskumar Pujari; Anshul Goel; Ashwin Sharma
International Journal Science and Technology Vol. 3 No. 3 (2024): November: International Journal Science and Technology
Publisher : Asosiasi Dosen Muda Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56127/ijst.v3i3.1968

Abstract

AI is increasingly used in high-stakes fields such as healthcare, finance, education, and public governance, requiring systems that uphold fairness, accountability, transparency, and privacy. This paper highlights the critical role of Supervised Fine-Tuning (SFT) in aligning large AI models with ethical principles and regulatory frameworks like the GDPR and EU AI Act. The interdisciplinary approach combines regulatory analysis, technical research, and case studies. It proposes integrating privacy-preserving techniques—differential privacy, secure multiparty computation, and federated learning—with SFT during deployment. The research also advocates incorporating Human-in-the-Loop (HITL) and Explainable AI (XAI) to ensure ongoing oversight and interpretability. SFT is positioned not only as a technical method but as a core enabler of responsible AI governance and public trust.

Page 1 of 1 | Total Record : 8