Claim Missing Document
Check
Articles

Found 31 Documents
Search

The Evolution of Image Captioning Models: Trends, Techniques, and Future Challenges Bastian, Ade; Wahid, Abrar; Hafsari, Zacky; Mardiana, Ardi
Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control Vol. 10, No. 4, November 2025
Publisher : Universitas Muhammadiyah Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22219/kinetik.v10i4.2305

Abstract

This study provides a comprehensive systematic literature review (SLR) of the evolution of image captioning models from 2017 to 2025, with a particular emphasis on the impending problems, methodological enhancements, and significant architectural developments. The evaluation is guided by the increasing demand for precise and contextually aware image descriptions, and it adheres to the PRISMA methodology. It selects 36 relevant papers from reputable scientific databases. The results indicate a significant transition from traditional CNN-RNN models to Transformer-based architectures, which leads to enhanced semantic coherence and contextual comprehension. Current methodologies, such as prompt engineering and GAN-based augmentation, have further facilitated generalization and diversity, while multimodal fusion solutions, which incorporate attention mechanisms and knowledge integration, have improved caption quality. Additionally, significant areas of concern include data bias, equity in model assessment, and support for low-resource languages. The study underscores the fact that modern vision-language models, such as Flamingo, GIT, and LLaVA, offer robust domain generalization through cross-modal learning and joint embedding. Furthermore, the efficacy of computing in restricted environments is improved by the development of pretraining procedures and lightweight models. This study contributes by identifying future prospects, analyzing technical trade-offs, and delineating research trends, particularly in sectors such as healthcare, construction, and inclusive AI. According to the results, in order to optimize their efficacy in real-world applications, future picture captioning models must prioritize resource efficiency, impartiality, and multilingual capabilities.
Run Length Encoding Compresion on Virtual Tour Campus to Enhance Load Access Performance Bastian, Ade; Mardiana, Ardi; Berliani, Mega; Firmansyah, Mochammad Bagasnanda
JOIN (Jurnal Online Informatika) Vol 8 No 1 (2023)
Publisher : Department of Informatics, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/join.v8i1.1000

Abstract

Virtual tour is one of the rapidly growing applications of multimedia technology which is used for various purposes, including the dissemination of information in an interesting way. The education sector is also not spared from using virtual tour media for promotional purposes, and campuses are no exception to this rule. Large virtual tour content causes high access speed, ultimately reducing the level of comfort experienced by users. This study aims to compress panoramic images displayed on a campus virtual tour using a lossless compression method and the Run Length Encoding (RLE) algorithm. First, panoramic images are combined into one, then individual images are compressed. When recreating a virtual campus tour, compressed images are used so that the amount of data transferred is smaller. The load access speed index increases from 7,233 seconds to 3,789 seconds when images are compressed from 64 bits to 8 bits, with a compression percentage of 27%. The findings from this research are that the RLE algorithm has not been able to compress large files effectively even though it is quite successful in increasing the load access of the virtual tour website.
RANCANG BANGUN APLIKASI PENDUKUNG PENGAMBILAN KEPUTUSAN PENYELEKSIAN CALON SISWA BARU Risma Miftahul Ulum, Ade; Mardiana, Ardi
SEMINAR TEKNOLOGI MAJALENGKA (STIMA) Vol 4 (2019): Inovasi Desa Menuju Era 4.0 : Merangsang Inovasi Desa Melalui Pemanfaatan Internet of
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The acceptance of new student is one of processes within educational institutions like school which is used to select candidate according to stated criteria by related school. Generally, the process of acceptance of new student began with registration, selection, and announcement steps of new student. This research was taken place at SMPN 1 Kasokandel which during the time has practiced the acceptance of new student manually or computerized but used Microsoft Office Excel only. It caused lack data, mistake within measurement because of manually mark measurement. Fundamental problem of this research is how to install information system for acceptance of new student at SMPN 1 Kasokandel that make information representation accurately and efficiently. therefore the author intends to create a new application-based student admission information system using the java programming language and MySQL database.
PERANCANGAN WEBSITE JURNAL DAN LAPORAN KULIAH KERJA NYATA MAHASISWA (KNM) UNIVERSITAS MAJALENGKA TAHUN 2019 Ridwan, Nugraha Sugiana; Mardiana, Ardi
SEMINAR TEKNOLOGI MAJALENGKA (STIMA) Vol 5 (2021): STIMA 5.0 Kebangkitan Nasional Digital Era Industri 4.0
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Student Real Work Lecture (KNM) is one form of devotion performed by students by giving lessons to the community in accordance with the knowledge that students have in their respective fields. During knm activities, students develop work programs that suit the needs and characteristics of the KNM implementation site accompanied by a report of work programs / KNM activities. The form of report that is still in the form of paper (manual) is considered ineffective because of limited storage space, limited availability of reports (report stock), and limited access to reports (which can only be accessed during business hours). This research was conducted with the aim to build an application that can provide convenience for students in processing KNM reports and reporting every activity during the KNM. This research study used the Rational Unified Process (RUP) and UML software development methodologies as software development models. Because the purpose in the beginning is to provide the availability of report access, the application that will be created is a web-based database application by utilizing Codeigniter Framework tools. From the research, a web-based application will be produced that can be accessed by each KNM participant.
ANALISIS PERAMALAN PENJUALAN MENGGUNAKAN METODE EXPONENTIAL SMOOTHING Mardiana, Ardi; Wulan Dari, Tiara
SEMINAR TEKNOLOGI MAJALENGKA (STIMA) Vol 7 (2023): STIMA 7.0 "Transformasi Peradaban Akademik Menyongsong Revolusi Industri 5.0"
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31949/stima.v7i0.847

Abstract

Pepro Cloth adalah perusahaan yang bergerak di bidang jasa pembuatan kaos, pdh, sweater, topi, mug, pin. Dalam mempersiapkan stok barang, Pepro Cloth masih dilakukan secara manual tanpa memperhitungkan barang terjual pada periode sebelumnya yang mengakibatkan kelebihan dan kekurangan stok barang. Hal tersebut dapat merugikan dan menghambat keuntungan bagi perusahaan. Oleh karena itu dibutuhkan suatu peramalan yang diharapkan dapat membantu perusahaan dalam menentukan jumlah stok barang yang akan dipesan pada periode selanjutnya. Metode peramalan yang digunakan adalah Metode Exponential Smoothing dengan alpha = 0,5, dan MAD, MSE dan MAPE sebagai standar error. Data yang digunakan diambil dari data stok barang pada tahun 2021-2022 sebagai referensi. Setelah dilakukan pengujian terhadap data tes enam bahan, dengan menggunakan metode Exponential Smoothing didapatkan hasil Periode selanjutnya bulan Agustus 2021, Yaitu Combed dengan tingkat akurasi 96,14%, Fleece dengan tingkat akurasi 90,86%, Drill dengan tingkat akurasi 96,18%, Topi dengan tingkat akurasi 92,25%, Mug dengan tingkat akurasi 91,5%, Pin dengan tingkat akurasi 79,69%. Dengan nilai standar error tersebut dapat simpulkan bahwa peramalan ini layak dan dapat diterapkan pada perusaan Pepro Cloth
PENGEMBANGAN DESAIN USER EXPERIENCE (UX) PADA WEBSITE KECAP SEGI TIGA MAJALENGKA Permana, Sri Mentari Widya Ningrum; Mardiana, Ardi
SEMINAR TEKNOLOGI MAJALENGKA (STIMA) Vol 9 (2025): Seminar Teknologi Majalengka (STIMA) 9.0 Tahun 2025
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The rapid development of digital technology has encouraged companies to utilize websites as a medium for promotion and branding. This study focuses on the development of user experience (UX) and user interface (UI) design for the Segi Tiga Soy Sauce website, which previously had limited digital presence and promotional reach. The objective of this research is to design a responsive, modern, and user-friendly website that improves customer engagement and supports the company’s business growth. The research applies the Design Sprint method, consisting of five stages: understand, diverge, decide, prototype, and validate. Data collection was conducted through interviews, observation, and literature studies to identify user needs and business requirements. The design process produced wireframes, storyboards, and prototypes using Figma, followed by usability testing with 20 respondents. The evaluation results using the System Usability Scale (SUS) showed a score of 87, indicating high user satisfaction and effective usability. The findings demonstrate that the proposed design not only enhances branding and professionalism but also improves market reach and communication efficiency for the company. This research highlights the importance of integrating UX principles in website development to achieve both functional and aesthetic goals in digital business promotion.
RANCANG BANGUN MODEL PENGENALAN EMOSI SUARA UNTUK PROMOSI PARIWISATA DI MAJALENGKA Rifki, Muhammad; Bastian , Ade; Mardiana, Ardi
SEMINAR TEKNOLOGI MAJALENGKA (STIMA) Vol 9 (2025): Seminar Teknologi Majalengka (STIMA) 9.0 Tahun 2025
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Emotion plays a crucial role in human communication, and recognizing emotions from speech (Speech Emotion Recognition or SER) has broad applications in various fields. This study designs and implements a speech emotion recognition application to support tourism promotion events in Majalengka. A 1D Convolutional Neural Network (CNN) model is developed using public SER datasets (RAVDESS, CREMA-D, SAVEE, TESS) combined and augmented to improve cultural generalization. Key audio features, such as Mel Frequency Cepstral Coefficients (MFCC), are extracted for effective emotion classification. The resulting system achieves an accuracy of 74.4% on test data, successfully recognizing emotions like angry, sad, neutral, and happy with good precision. This automated emotion analysis assists judges in evaluating participants’ speeches objectively and efficiently. The integration of SER technology in tourism events demonstrates an innovative strategy to enhance the promotion of local culture and improve the overall visitor experience in Majalengka.
PENGISIAN BAHAN BAKAR SELF-SERVICE DENGAN PEMBAYARAN NON TUNAI MELALUI ANIMASI 3D BLENDER Adriansyah, Rizky Adin; Mardiana, Ardi
SEMINAR TEKNOLOGI MAJALENGKA (STIMA) Vol 9 (2025): Seminar Teknologi Majalengka (STIMA) 9.0 Tahun 2025
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

This study focuses on the development of a 3D animation as an educational medium for the selfservice fuel filling system integrated with cashless payment methods at gas stations. The research was motivated by the limited adoption of self-service systems in Indonesia and the increasing need for safe, convenient, and contactless transactions, especially after the COVID-19 pandemic. The study employed the Multimedia Development Life Cycle (MDLC) method consisting of six stages: concept, design, material collecting, assembly, testing, and distribution. The testing phase was carried out in two stages: Black box Testing to evaluate functionality and usability testing with 40 respondents using a questionnaire. The results showed that the developed animation achieved user satisfaction levels of 76.75% for ease of understanding, 75.00% for cashless payment integration, and 82.50% for effectiveness as an educational tool, with an overall average of 77.84%. These findings indicate that the 3D animation is feasible and effective as a learning medium to improve user understanding of self-service fuel filling with cashless payment at gas stations.
PERANCANGAN SISTEM INFORMASI MANAJEMEN DISTRIBUSI AGEN GAS ELPIJI BERBASIS WEB PADA PT. BAROKAH TIGA PUTRA Kiranti, Mei Bunga; Mardiana, Ardi
SEMINAR TEKNOLOGI MAJALENGKA (STIMA) Vol 9 (2025): Seminar Teknologi Majalengka (STIMA) 9.0 Tahun 2025
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Barokah Tiga Putra is an LPG (Liquefied Petroleum Gas) agent in Majalengka Regency that distributes 3-kilogram LPG cylinders to more than 60 distribution bases. The current manual distribution process using Microsoft Excel leads to several problems, such as slow data access, recording errors, and the inability to monitor distribution in real time. This research aims to design and implement a web-based distribution management information system to improve efficiency, accuracy, and operational effectiveness. The system development method applied is Rapid Application Development (RAD), which consists of requirement planning, design, testing, and implementation stages. The implementation results show that the system successfully facilitates the management of base data, requests, deliveries, sales, and LPG receipts in an integrated manner. Furthermore, the system enables real-time monitoring of distribution, supporting agents in making strategic decisions. Therefore, the developed information system provides a practical solution to overcome the limitations of the previous manual distribution process.
RANCANG BANGUN WEBSITE KETERSEDIAAN OBAT UNTUK MENINGKATKAN AKSES DAN LAYANAN KESEHATAN Amelia, Rizki; Mardiana, Ardi
SEMINAR TEKNOLOGI MAJALENGKA (STIMA) Vol 9 (2025): Seminar Teknologi Majalengka (STIMA) 9.0 Tahun 2025
Publisher : Universitas Majalengka

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The healthcare industry requires an efficient drug inventory management system to improve service quality and patient access to medicines. This research aims to develop a web-based drug availability application using the Rapid Application Development (RAD) methodology. The system includes drug inventory management, patient data management, transaction recording (incoming and outgoing drugs), and automatic reporting features. This web-based application was developed using NextJS framework, JavaScript programming language, and MySQL database to provide a user-friendly interface for clinic staff. The development methodology follows the RAD approach which consists of requirements planning, user design, construction, and cutover phases. Black box testing results show that all system functions work properly with 100% success rate. The system successfully provides automated drug inventory management, transaction monitoring, and comprehensive reporting features that can improve operational efficiency at Guvili Clinic. This drug availability management system offers an effective solution to replace manual recording methods and provides real-time access to drug inventory data for better healthcare service management.