cover
Contact Name
Sanriomi Sintaro
Contact Email
techcartpress@gmail.com
Phone
+6282180318941
Journal Mail Official
jaiti.techcartpress@gmail.com
Editorial Address
Jalan Gatot Subroto No 47 LK I, Desa/Kelurahan Tanjunggading, Kec. Kedamaian, Kota Bandar Lampung, Provinsi Lampung, Indonesia
Location
Kota bandar lampung,
Lampung
INDONESIA
Journal of Artificial Intelligence and Technology Information
Published by Tech Cart Press
ISSN : 29855306     EISSN : 29856396     DOI : doi.org/10.58602/jaiti
Journal of Artificial Intelligence and Technology Information (JAITI) is a peer-review journal focusing on Artificial Intelligence and Technology Information issues. Journal of Artificial Intelligence and Technology Information (JAITI) invites academics and researchers who do original research in artificial intelligence and technology 1nfromation.Journal of Artificial Intelligence and Technology Information (JAITI) is published by Tech Cart Press in March, June, September, and December every year. Journal of Artificial Intelligence and Technology Information (JAITI) accept articles in Bahasa Indonesia and English.
Articles 114 Documents
Systematic Literature Review: Integration of WebQual 4.0, Citizen Trust, and User Acceptance Testing in the Analysis of Government Website Quality Rinda Hesti Kusumaningtyas; Ikaningrum Nurazizah Ningrum; A'ang Subiyakto
Journal of Artificial Intelligence and Technology Information (JAITI) Vol. 4 No. 4 (2026): Volume 4 Number 4 December 2026 (Online First)
Publisher : PT. Tech Cart Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58602/jaiti.v4i4.427

Abstract

Digital transformation in the government sector has driven increased use of websites to provide information and public services. However, the mere existence of a government website does not automatically imply that the service meets the public's needs and expectations. Website quality and user trust are critical aspects to consider when developing government digital services. This study aims to examine the application of WebQual 4.0 and Citizen Trust in evaluating government website quality and to formulate an improvement process via a redesign prototype, which is subsequently tested using User Acceptance Testing (UAT). The research employs a Systematic Literature Review (SLR) approach, guided by PRISMA protocols, to identify, select, and synthesize relevant prior studies. The analysis focuses on the three WebQual 4.0 dimensions of usability, information quality, and service interaction quality, alongside the concept of Citizen Trust within the context of government digital services. Findings from the literature synthesis are used to identify quality and trust aspects requiring improvement, serving as the basis for developing the redesign prototype. Subsequently, UAT is utilized to assess the extent to which the resulting prototype aligns with user needs and is accepted by users. This research aims to produce an evaluation framework linking website quality, public trust, the redesign process, and user acceptance testing, thereby offering a systematic approach to enhancing the quality of government digital services.
Deteksi Berita Hoaks Berbahasa Indonesia Menggunakan Artificial Intelligence Berbasis Natural Language Processing dan IndoBERT Jonandes Anthony Sinaga; Hendi Suhendi
Journal of Artificial Intelligence and Technology Information (JAITI) Vol. 4 No. 4 (2026): Volume 4 Number 4 December 2026 (Online First)
Publisher : PT. Tech Cart Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58602/jaiti.v4i4.428

Abstract

Penyebaran teknologi informasi media digital pada masyarakat sudah meningkat sangat pesat dan mempercepat penacarian sebuah informasi. Tetapi, kemudahan yang tersedia juga memunculkan penyebaran berita hoaks yang sangat tinggi sehingga berpotensi menimbulkan disinformasi, kepanikan, serta konflik sosial pada masyarakat. Proses verifikasi berita secara manual masih memiliki keterbatasan karena membutuhkan waktu yang cukup lama. Oleh karena itu, penelitian ini dibuat untuk mengembangkan sistem pendeteksi berita hoaks berbahasa Indonesia dengan memanfaatkan teknologi Artificial Intelligence (AI) melalui pendekatan Natural Language Processing (NLP) dan model IndoBERT. Penelitian ini dilakukan menggunakan dataset berita hoaks dan berita valid yang dikumpulkan dari berbagai sumber terpercaya yang ada di Indonesia selama 5 tahun terakhir. Tahapan penelitian meliputi pengumpulan data, preprocessing teks yang terdiri atas cleaning, case folding, tokenization, stopword removal, dan stemming, lalu dengan proses klasifikasi menggunakan model IndoBERT. Kinerja model dievaluasi menggunakan metrik Accuracy, Precision, Recall, F1-Score, serta Confusion Matrix. Penerapan model berbasis transformer IndoBERT diharapkan mampu memahami konteks bahasa Indonesia secara lebih baik dibandingkan metode klasifikasi konvensional sehingga menghasilkan performa deteksi yang lebih akurat. Hasil penelitian ini diharapkan dapat menjadi alternatif solusi dalam mendukung proses identifikasi berita hoaks secara otomatis, meningkatkan efisiensi kepastian informasi, serta memberikan kontribusi terhadap penerapan Artificial Intelligence pada bidang Sistem Informasi.
Formulation of a Weighted Scoring Model Based on Forensic Medical Examination Data as a Decision Support for Prioritizing Sexual Violence Cases Syifaurachman Syifaurachman; Stefanus Satryo Wasono; Ragil Nur Iman; Yosia Simbolon
Journal of Artificial Intelligence and Technology Information (JAITI) Vol. 4 No. 4 (2026): Volume 4 Number 4 December 2026 (Online First)
Publisher : PT. Tech Cart Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58602/jaiti.v4i4.436

Abstract

Recent empirical studies highlight that post-factum clinical forensic records in sexual violence cases remain underutilized, leaving emergency case handling heavily dependent on intuitive or first-come, first-served administrative practices that delay critical interventions. This study aims to formulate and validate an automated decision support model utilizing the Simple Additive Weighting (SAW) algorithm to structure medicolegal data into an objective priority-ranking mechanism for case handling. Practically, this framework enables public hospitals and victim protection units to optimize scarce emergency resources, safeguard golden hours for medical prophylaxis, and execute immediate safe-house evacuations. The research method extracted five clinical-forensic criteria—victim age vulnerability, physical and or genital trauma severity, perpetrator relationship, reporting delay, and toxicological or weapon threats from 50 de-identified clinical forensic records, normalized through linear cost-benefit functions weighted by forensic specialists. The results demonstrate robust algorithmic stability across varying weight scenarios, with Spearman's rank correlation reaching and , alongside an empirical expert concordance of () that successfully categorized 22% of cases into Emergency Handling Priority I. The research completely meets the expected computational precision and validity, confirming that the model provides a transparent handling prioritization mechanism. However, further multi center trials and integration with natural language processing tools are needed to automate textual feature extraction in larger hospital settings.
Design and Development of a Web-Based Library Information System at SMP Negeri 221 Jakarta Muhammad Aditya Firdaus; Tuti Haryanti
Journal of Artificial Intelligence and Technology Information (JAITI) Vol. 4 No. 4 (2026): Volume 4 Number 4 December 2026 (Online First)
Publisher : PT. Tech Cart Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58602/jaiti.v4i4.438

Abstract

The library administration at SMP Negeri 221 Jakarta previously relied on conventional ledger-based records. This practice created difficulties in retrieving data, increased the risk of recording errors and duplication, and delayed report preparation. This study aims to design and develop a web-based library information system to support the management of student records, books, categories, borrowing and returning transactions, user accounts, and borrowing reports. The system was developed using the Rapid Application Development (RAD) method, consisting of requirements planning, user design, construction, and cutover/implementation. Data were collected through observation, interviews, and literature study. The application was built using Laravel, PHP, MySQL, Blade Template Engine, and Bootstrap. Black Box Testing was conducted based on the functional scenarios defined during system design. The testing results achieved a 100% success rate, indicating that all tested scenarios—including user authentication, student and book data management, circulation transactions, and report generation—operated according to the expected results. The system also provides user activity logging and late-return fine calculation. Therefore, the developed system can support the digitalization of library administration by integrating essential management functions into a centralized platform. Future improvements may include self-service borrowing for students, return-date reminders, and more comprehensive analytical reporting features.

Page 12 of 12 | Total Record : 114