Article Per Year (5 Year)
p-Index From 2021 - 2026
3.58
P-Index
This Author published in this journals
All Journal REKA KARSA JURNAL FOKUS KONSELING Register: Jurnal Ilmiah Teknologi Sistem Informasi JURNAL KEPEMIMPINAN DAN PENGURUSAN SEKOLAH JIPI (Jurnal Ilmiah Penelitian dan Pembelajaran Informatika) Al-Syakhshiyyah : Jurnal Hukum Keluarga Islam dan Kemanusiaan Jurnal FASILKOM (teknologi inFormASi dan ILmu KOMputer) KomunikA Jurnal Pengabdian Masyarakat Bhinneka Jurnal Pengabdian Masyarakat Bangsa PPSDP International Journal of Education Horizon: Indonesian Journal of Multidisciplinary Malacca: Journal of Management and Business Development
Agung Prasetya
Jurusan Teknik Arsitektur Fakultas Teknik Sipil dan Perencanaan Institut Teknologi Nasional
Religion Agriculture, Biological Sciences & Forestry Arts Humanities Civil Engineering, Building, Construction & Architecture Computer Science & IT Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Energy Health Professions Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Mathematics Medicine & Pharmacology Nursing Physics Public Health Social Sciences Other

Published : 30 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 30 Documents
Search

 1   2   3 
SQL Structure Embedding Berbasis Transformer untuk Perankingan Few-Shot Example pada Sistem Text-To-SQL Pamudyo, Argo Yanuar; Prasetya, Agung
HORIZON: Indonesian Journal of Multidisciplinary Vol. 4 No. 4 (2026): HORIZON: Indonesian Journal of Multidisciplinary
Publisher : Lembaga Intelektual Muda (LIM) Maluku

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54373/hijm.v4i4.6834

Abstract

This research develops and evaluates a Transformer encoder-based SQL Structure Embedding (SSE) to improve SQL structure representation quality in Text-to-SQL systems. Static embedding approaches such as GloVe, commonly used in few-shot learning-based Large Language Model (LLM) systems, have limited ability to capture inter-clause contextual relationships in SQL, thereby reducing the quality of few-shot example selection. The dataset used is the BIRD dev set with 1,534 Natural Language Query (NLQ) and SQL query pairs, annotated with five SQL latent intent categories and 44 SQL component labels. The SSE model is built using an encoder-only Transformer architecture initialized with 100-dimensional pretrained GloVe weights, two encoder layers with 4 attention heads, and mean pooling. Evaluation results show the SSE model achieves a Micro F1 of 0.8671, Macro F1 of 0.7920, and Samples F1 of 0.8312, improving by 9.17%, 10.31%, and 9.20% respectively over the GloVe baseline. The combination of SSE with a downstream classifier yields the best performance with a Micro F1 of 0.8821. These results confirm that the self-attention mechanism in Transformer more effectively captures relational context between SQL clauses to support few-shot example ranking in Text-to-SQL systems.
Deteksi Ill-Formed Natural Language Query Berbasis Rule Pada Model Llm Text-to-SQL Berbahasa Indonesia Mukti, Ricky Fajar; Prasetya, Agung; Cahyono, Taufiq Agung
HORIZON: Indonesian Journal of Multidisciplinary Vol. 4 No. 4 (2026): HORIZON: Indonesian Journal of Multidisciplinary
Publisher : Lembaga Intelektual Muda (LIM) Maluku

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54373/hijm.v4i4.6916

Abstract

The performance of a Large Language Model (LLM)-based Text-to-SQL system is heavily influenced by the quality of the user-supplied Natural Language Queries (NLQs). Ill-formed NLQs, such as incomplete, ambiguous, or inconsistent information within the database schema, can generate syntactically valid but semantically incorrect SQL queries. This study aims to develop a rule-based ill-formed NLQ detection system as a pre-processing stage in an Indonesian Text-to-SQL pipeline. The system is built using the Experta library with 15 IF-THEN rules that identify six categories of ill-formed NLQs: insufficient information, multiple interpretations, incomplete conditions, illogical, out-of-schema reference, and unsafe/non-SELECT. After rule-only filtering, a LLaMA 3.2 3B Instruct-based validator is optionally used to examine cases requiring deeper semantic understanding. An evaluation of 202 Indonesian NLQs showed that the Rule-only mode achieved 89.60% Recall, 78.87% Precision, 83.90% F1-Score, and 78.71% Accuracy. False positives were primarily influenced by overly strict rules, while false negatives were dominated by semantic ambiguity. These results demonstrate that the rule-based approach provides a consistent and transparent initial filtering mechanism and can be combined with an LLM-based validator to handle cases requiring more complex semantic interpretation.  
Rancang Bangun Sistem Informasi Akademik Berbasis Website Menggunakan Metode Design Science Research Methodology (Studi Kasus: SMK Veteran 1 Tulungagung) Sanggara, Arganata; Cahyono, Taufik Agung; Prasetya, Agung
HORIZON: Indonesian Journal of Multidisciplinary Vol. 4 No. 4 (2026): HORIZON: Indonesian Journal of Multidisciplinary
Publisher : Lembaga Intelektual Muda (LIM) Maluku

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54373/hijm.v4i4.6930

Abstract

Academic data management at SMK Veteran 1 Tulungagung still faces obstacles in the form of manual administrative processes, unintegrated data, reporting delays, and limited access to real-time information. These conditions indicate the need to develop an academic information system capable of supporting more effective and concise data management. This study aims to develop a website-based Academic Information System (SIAKAD) to improve the efficiency of academic administration management in the school environment. The research method uses the Design Science Research Methodology (DSRM) approach through the stages of problem identification, design, development, implementation, and system evaluation. The results of functional testing using Black Box Testing show that all features run as needed with a 100% success rate. Usability evaluation using the System Usability Scale (SUS) obtained an average score of 75.23 which is included in the acceptable category with a good level of user acceptance. In addition, performance testing shows that the system has a fast page loading time with a Largest Contentful Paint (LCP) value of under 1 second. The results of the study indicate that the developed SIAKAD is able to integrate academic data management in a single opening platform, simplifying information access, and supporting increased effectiveness of educational administration in vocational schools
Pengembangan Sistem Informasi Pemesanan dan Reservasi Restoran Alam Kitchen Tulungagung Berbasis Website Menggunakan Metode Feature Driven Development (FDD) Gunawan, Rafli Nur; CahyonoCahyono, Taufiq Agung; Prasetya, Agung
HORIZON: Indonesian Journal of Multidisciplinary Vol. 4 No. 4 (2026): HORIZON: Indonesian Journal of Multidisciplinary
Publisher : Lembaga Intelektual Muda (LIM) Maluku

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54373/hijm.v4i4.6931

Abstract

Alam Kitchen Restaurant in Tulungagung still uses a manual ordering and reservation process via telephone and WhatsApp, which causes risks of recording errors, service delays, and limitations in monitoring table availability in real time. This study aims to develop a website-based restaurant ordering and reservation information system and evaluate the feasibility of the resulting system. The study uses a Research and Development (R&D) method with a Feature Driven Development (FDD) approach that includes the stages of model design, feature identification, planning, design, and system feature development. Data collection was carried out through observation, interviews, and literature studies to identify user needs. System evaluation was carried out through functional testing using Black Box Testing and User Acceptance Testing (UAT) on 40 respondents. The test results showed that all system features run according to functional requirements, while the UAT results obtained a feasibility percentage of 85.75% with the category "Very Feasible". These findings indicate that the developed system is able to support the ordering process, table reservations, payment transactions, and restaurant data management in a more structured manner and improve service effectiveness compared to the previous manual system
Pembangunan Multi-Domain Human-Labelled Dataset untuk Konversi Bahasa Alami ke SQL pada Bahasa Indonesia Firlanda, Fiorella Asyfa; Prasetya, Agung
JURNAL FASILKOM Vol. 16 No. 2 (2026): Jurnal FASILKOM (teknologi inFormASi dan ILmu KOMputer)
Publisher : Unversitas Muhammadiyah Riau

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37859/jf.v16i2.12020

Abstract

This study develops a multi-domain, human-labelled Text-to-SQL dataset in Indonesian to address the absence of such resources for the language. The dataset was constructed through a corpus-based approach comprising four stages: corpus planning, corpus construction, corpus validation, and model evaluation. Database schemas represented as Entity Relationship Diagrams (ERD) were used as the structural foundation for generating pairs of Indonesian natural language questions and SELECT-type SQL queries, all created through manual annotation. Validation was performed through four sequential mechanisms: SQL syntax checking, schema conformity verification, query execution testing, and semantic alignment assessment between questions and SQL queries. The resulting dataset is Spider-compatible in JSON format, covering 40 databases, 27 domains, 1,524 questions, and 1,355 unique SQL queries distributed across four difficulty levels. Preliminary evaluation using SQLNet and TypeSQL baseline models under example split and database split scenarios confirms that the dataset provides a representative and challenging evaluation environment for Indonesian Text-to-SQL experiments, though model performance remains limited on complex queries and previously unseen database schemas. The dataset is publicly available and is intended to support future development and evaluation of Indonesian Text-to-SQL models..
Pengembangan Aplikasi GPS Tracking Dan Laporan Digital untuk Mendukung Monitoring Personel Lapangan di Polres Tulungagung Ardiansyah, Diva Yudis; Cahyono, Taufiq Agung; Prasetya, Agung
HORIZON: Indonesian Journal of Multidisciplinary Vol. 4 No. 3 (2026): HORIZON: Indonesian Journal of Multidisciplinary
Publisher : Lembaga Intelektual Muda (LIM) Maluku

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54373/mn9whd36

Abstract

Penelitian ini bertujuan mengembangkan Aplikasi GPS Tracking dan Laporan Digital yang dapat digunakan secara operasional di lingkungan Polres Tulungagung. Metode yang digunakan adalah Research and Development (R&D) dengan model pengembangan ADDIE (Analysis, Design, Development, Implementation, Evaluation). Sistem yang dikembangkan terdiri dari dua komponen utama, yaitu dashboard web berbasis Laravel untuk administrator dan aplikasi mobile berbasis Flutter untuk personel lapangan. Integrasi real-time diwujudkan melalui Pusher WebSocket dengan mekanisme fallback polling REST API, serta algoritma Dead Reckoning berbasis formula Haversine untuk menghasilkan animasi pergerakan marker yang mulus pada render loop 60 FPS. Fitur utama sistem meliputi GPS tracking real-time dengan visualisasi peta interaktif menggunakan Leaflet.js dan flutter_map, manajemen instruksi dan jadwal patroli, pelaporan digital dengan lampiran foto dan video, checkpoint lokasi, ekspor laporan PDF menggunakan DomPDF, serta autentikasi berbasis Role-Based Access Control (RBAC). Background service foreground Android memastikan pelacakan GPS tetap berjalan meskipun aplikasi diminimisasi. Hasil implementasi menunjukkan seluruh fitur yang direncanakan berhasil dikembangkan dan berfungsi sesuai kebutuhan operasional Polres Tulungagung.
Identifikasi Kuantitas Relevan pada Soal Cerita Matematika Menggunakan Model BiLSTM dan CRF Hidayat, Muhamad Taufiq; Prasetya, Agung; Iskandar, Joko
HORIZON: Indonesian Journal of Multidisciplinary Vol. 4 No. 4 (2026): HORIZON: Indonesian Journal of Multidisciplinary
Publisher : Lembaga Intelektual Muda (LIM) Maluku

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54373/3fj09730

Abstract

Mathematical word problems often contain information that is not entirely relevant to answering the question, causing students to make errors in selecting important information. This study aims to identify relevant quantities in Indonesian mathematical word problem texts using a sequence labeling approach. The dataset consists of 700 problems collected from elementary school mathematics e-books and annotated using the BIO scheme (B-REL, I-REL, and O). The model employed is a combination of BiLSTM and CRF with an 80:20 train–test split. Evaluation is conducted using precision, recall, F1-score, and accuracy, with a focus on the F1-score of the REL class. The results show that the model achieves an F1-score of 0.9726. The dominant error occurs in the misclassification of REL as O, indicating that quantity spans are often truncated across tokens. However, no violations of the BIO scheme are observed, ensuring consistent label sequences. These findings indicate that the combination of BiLSTM and CRF is effective in identifying relevant quantities and has the potential to support the understanding of mathematical word problems.
Identification of Quantity Relationships in Math Story Problems Using Bidirectional Long Short-Term Memory (Bi-LSTM) Puspita, Diana Ayu; Prasetya, Agung; Sari, Yayak Kartika
HORIZON: Indonesian Journal of Multidisciplinary Vol. 4 No. 3 (2026): HORIZON: Indonesian Journal of Multidisciplinary
Publisher : Lembaga Intelektual Muda (LIM) Maluku

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54373/hijm.v4i3.6049

Abstract

Penelitian ini bertujuan untuk mengidentifikasi hubungan kuantitas dalam soal cerita matematika menggunakan metode Bidirectional Long Short-Term Memory (BiLSTM). Permasalahan dalam penelitian ini adalah kesulitan dalam memahami dan mengklasifikasikan hubungan kuantitas secara otomatis akibat variasi struktur kalimat dan konteks bahasa. Metode yang digunakan adalah pendekatan deep learning dengan tahapan pengumpulan data, preprocessing, pembentukan vocabulary, pelatihan model, dan evaluasi model. Hasil penelitian menunjukkan bahwa model BiLSTM mampu mengklasifikasikan hubungan kuantitas dengan baik dengan memperoleh nilai accuracy sebesar 0.9224 serta nilai precision, recall, dan F1-score yang relatif tinggi pada setiap kelas. Hal ini menunjukkan bahwa model memiliki performa yang baik dalam memahami pola hubungan kuantitas dalam teks soal cerita matematika, sehingga dapat digunakan sebagai pendekatan yang efektif untuk identifikasi hubungan kuantitas secara otomatis.
Rancang Bangun Sistem Informasi Manajemen Administrasi pada Bnn Kabupaten Trenggalek dengan Metode Feature-Driven Development Safa’a, Tri Anandhya; Cahyono, Taufiq Agung; Prasetya, Agung
HORIZON: Indonesian Journal of Multidisciplinary Vol. 4 No. 4 (2026): HORIZON: Indonesian Journal of Multidisciplinary
Publisher : Lembaga Intelektual Muda (LIM) Maluku

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54373/hijm.v4i4.6830

Abstract

Administrative management at the Trenggalek Regency National Narcotics Agency (BNN) is currently handled manually using physical logbooks and Microsoft Excel, resulting in inefficient administrative processes, a lack of data integration, and suboptimal document disposition workflows. This study aims to design and develop a web-based administrative management information system using the Feature-Driven Development (FDD) method. The system was developed using PHP and the Laravel framework, incorporating features for managing incoming and outgoing correspondence, official orders, document dispositions, a digital guestbook, and vehicle reports. System testing was conducted using Black Box Testing and the System Usability Scale (SUS). Black Box Testing results indicated that all 16 test scenarios for the system's core features were successful, achieving a 100% success rate. These tests covered login and validation processes, correspondence management, official orders, dispositions, the digital guestbook, vehicle reports, and date-based data filtering. Meanwhile, the SUS evaluation yielded a score of 76.5—categorized as "OK" with a grade of C—which exceeds the benchmark score of 68. Consequently, the system successfully performs all designed functions and demonstrates a good level of user acceptance, making it suitable for supporting integrated and computerized administrative management at BNN Trenggalek Regency.
Generator Soal Cerita Kurikulum Matematika Dasar Berbasis Large Language Model Rohadi, Jalmo; Prasetya, Agung; Ansor, Mohamad Khoirul
HORIZON: Indonesian Journal of Multidisciplinary Vol. 4 No. 4 (2026): HORIZON: Indonesian Journal of Multidisciplinary
Publisher : Lembaga Intelektual Muda (LIM) Maluku

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54373/hijm.v4i4.7369

Abstract

Math word problems (MWPs) are essential for developing the reasoning skills of elementary school students; however, manual creation is time-consuming and results in limited problem variety. This study aims to design, implement, and evaluate a Large Language Model (LLM)-based math word problem generator that aligns with the Kurikulum Merdeka (Independent Curriculum) and produces valid, solvable problems in Indonesian. The system was developed using LLaMA 3.2 3B Instruct via QLoRA fine-tuning with 4-bit NormalFloat quantization, trained on a dataset of 3,185 entries covering 19 topics and 6 grade levels. The system generated 385 problems across 77 grade-topic combinations and was evaluated against a sample of 95 problems through assessments by three elementary school teachers and the G-Eval metric. Evaluation results indicate that 93 out of 95 problems met validity criteria (97.89%), with an average quality score of 4.95 out of 5.00. A comparison of the two evaluation methods revealed a Pearson correlation of 0.94 and a Cohen’s Kappa of 1.00. These findings demonstrate the system's potential as a tool to assist teachers in generating math word problems that align with the Kurikulum Merdeka.
Co-Authors Ansor, Mohamad Khoirul Anwar, Ndaru Syaiful Ardiansyah, Diva Yudis Cahyono, Taufik Agung CahyonoCahyono, Taufiq Agung Chastine Fatichah Dahnilsyah, Dahnilsyah Dessy Wardiah, Dessy Desy Arum Sunarta Fadly Azhar Fahrezi, Dion Maulana Faturrahim, Muhammad Firlanda, Fiorella Asyfa Gunawan, Rafli Nur Harry Perdana Hidayat, Muhamad Taufiq Ilham Arvan Junaidi Iskandar, Joko Mahdum Minanurohman, Muhamad Adib Mukti, Ricky Fajar Nur Laela Latifah Nurul Aini Rahmawati Oswald P. M Siahaan Pamudyo, Argo Yanuar Puspita, Diana Ayu Putra, Valentino Eka Apritha Rahmawati, Rahmawati Ramadhaningsih, Dwi Rambe, Armita Novriana Riski Apriliani Rohadi, Jalmo Rusdaya Basri, Rusdaya Rusdianto Rusdianto, Rusdianto Safa’a, Tri Anandhya Sanggara, Arganata Sari, Nurul Permata Sari, Yayak Kartika Sunuwati, Sunuwati Syafi', Faiza Taufiq Agung Cahyono Umi Laili Yuhana