Jurnal Ilmiah Media Sisfo

Analisis Kepuasan Pengguna Meta AI Instagram Dengan Model TAM-EUCS

Priskha Trivenna — 2025-10-31

Meta AI is a feature of artificial intelligence created by Meta Inc to assist users in generating content and acquiring information in an interactive manner. This research seeks to assess user satisfaction with the Meta AI feature by employing a combined methodology of the Technology Acceptance Model and End User Computing Satisfaction. The research included 403 participants who actively use Instagram and have utilized the Meta AI feature. The analysis technique employed was Partial Least Squares Structural Equation Modeling to investigate the impact of seven independent variables on user attitude and the effect of user attitude on user satisfaction. The findings indicated that perceived usefulness, perceived ease of use, and format significantly influence user attitude, whereas content, accuracy, technical ease, and timeliness do not show a notable impact. Moreover, user perception significantly impacts user satisfaction. Ultimately, perceived advantages, accessibility, and visual layout are the primary elements influencing user perceptions and contentment regarding AI-driven functionalities on social media.

Rancang Bangun Prototipe Teknologi Persuasif Untuk Pengembangan Website Perpustakaan Berbasis Sistem Rekomendasi

Miftah Anantha — 2025-10-31

The development of information technology has brought significant changes to various sectors of life, including the world of libraries. With the rapid innovation in technology, various aspects of library services have now been transformed thru digitalization, enabling faster access to information. Nevertheless, many libraries still rely on non-digital lending systems. This research aims to design and develop a persuasive technology prototype for the Bina Darma University library website, integrating a Collaborative Filtering-based recommendation system to provide relevant book recommendations based on user preferences. The research methods used include usability testing and the Perceived Persuasiveness Questionnaire (PPQ) to evaluate the effectiveness and efficiency of the system. The research results show that the system prototype successfully improved effectiveness, with a success rate of 97.5%, efficiency of 88%, and a System Usability Scale (SUS) score of 79, indicating that the system is user-friendly and effective in changing user behavior toward technology-based book borrowing

Audit Tata Kelola Teknologi Informasi Universitas Nurdin Hamzah Menggunakan Framework COBIT 2019

Elna Sahara — 2025-10-31

Teknologi Informasi (TI) merupakan komponen strategis yang berperan krusial dalam menunjang operasional serta pencapaian tujuan organisasi, termasuk di lingkungan perguruan tinggi. Tujuan penelitian ini dalam rangka mengevaluasi tata kelola TI di Universitas Nurdin Hamzah menggunakan kerangka kerja COBIT 2019, yang terdiri atas 40 proses dalam lima area utama, yaitu EDM, APO, BAI, DSS, dan MEA. Audit difokuskan pada dua domain, yakni BAI03 (Managed Solutions Identification and Build) dan BAI06 (Managed Changes). Pendekatan deskriptif kuantitatif digunakan pada penelitian ini memanfaatkan metode pengumpulan data lewat wawancara terstruktur, penyebaran kuesioner kepada unit pengelola TI, serta telaah dokumen kebijakan dan prosedur terkait. Hasil audit menunjukkan bahwa BAI03 pada capability level 3 (Established Process) yang memiliki capaian 75,8%, sedangkan BAI06 pada level 4 (Predictable Process) dengan capaian 70%, yang keduanya tergolong largely achieved. Beberapa indikator kapabilitas belum terpenuhi, seperti belum adanya pelibatan penguji independen, kurangnya dokumentasi status perubahan, serta belum adanya penetapan prioritas terhadap permintaan perubahan TI. Berdasarkan hasil tersebut, disusun rekomendasi perbaikan dalam upaya peningkatan kapabilitas setiap proses agar tata kelola TI dapat berjalan lebih efektif dan efisien. Penelitian ini memberi beberapa masukan strategis untuk peningkatan tata kelola TI di perguruan tinggi. Sementara secara teoretis, penelitian ini berkontribusi dalam memperkaya penerapan dan pengembangan COBIT 2019 di sektor pendidikan tinggi melalui adaptasi domain BAI terhadap karakteristik institusi akademik.

Analisis dan Perancangan Sistem Informasi SDMK Menggunakan Algoritma Random Forest Berbasis Web

mery — 2025-10-31

Sistem informasi yang terintegrasi kini menjadi kebutuhan utama bagi rumah sakit untuk meningkatkan efisiensi dan kualitas pelayanan. Rumah Sakit Umum Daerah Raden Mattaher Jambi menghadapi sejumlah permasalahan dalam mengelola data dan kompetensi tenaga kesehatan yang semakin kompleks, seperti kesulitan dalam pelacakan data kompetensi, ketidakakuratan dalam penilaian, dan keterlambatan dalam pengambilan keputusan manajerial. Permasalahan ini mengakibatkan pengelolaan sumber daya manusia menjadi kurang optimal. Penelitian ini bertujuan untuk menganalisis dan merancang Sistem Informasi SDMK (Sumber Daya Manusia Kesehatan) berbasis web dengan menggunakan algoritma Random Forest. Algoritma ini dipilih karena kemampuannya yang tinggi dalam mengolah data kompleks dan menghasilkan prediksi yang akurat. Metode penelitian ini meliputi analisis kebutuhan sistem melalui wawancara dan observasi, perancangan sistem menggunakan model Random Forest, dan pengembangan aplikasi berbasis web. Diharapkan hasil penelitian ini dapat menghasilkan sebuah sistem informasi yang tidak hanya meningkatkan akurasi dalam penilaian kompetensi tenaga kesehatan tetapi juga mempercepat proses pengambilan keputusan yang berbasis data. Dengan demikian, penelitian ini diharapkan dapat berkontribusi pada peningkatan kualitas manajemen sumber daya manusia di RSUD Raden Mattaher.

Analysis of the Success of SKPI Information System of the University of Jambi Using the DeLone and McLean Model

Reidila Ariska Widya — 2025-10-31

The University of Jambi as a higher education institution also prioritizes technological advances in its operations. In 2021, the Diploma Companion Certificate (SKPI) information system was developed, which was then officially released in 2022. This system functions to support data collection of student achievements and activities. However, until now, the SKPI information system is still in the refinement stage for the latest versions and features. This study aims to determine the success rate of the SKPI information system using the DeLone and McLean model. The approach used is quantitative, with response-based data collection and analysis techniques from students of the Faculty of Science and Technology, University of Jambi. The sampling method in this study uses the Isaac and Michael formula, with a sample of 100 respondents and an error rate of 10%. The distribution of the questionnaire was carried out online. The data obtained were analyzed using the Partial Least Square - Structural Equation Modeling (PLS-SEM) method, which included outer model testing, inner model, and hypothesis testing. The results showed that of the 9 hypotheses tested, 5 hypotheses had a positive and significant influence on the success of the SKPI information system, while the other 4 hypotheses did not show a significant influence.

Pengembangan Sistem Pendukung Keputusan Berbasis Web untuk Diagnosis Anemia dengan Metode Certainty Factor

Muhammad Nurwegiono — 2025-10-31

Adolescent girls are a vulnerable group prone to anemia due to increased iron requirements during puberty. Undiagnosed anemia can negatively impact health, productivity, and academic performance. This study aims to design and develop a web-based Decision Support System (DSS) to assist in diagnosing various types of anemia based on symptoms experienced by users, using the Certainty Factor (CF) method. The system was developed using the Waterfall model, which includes requirement analysis, system design, coding, and testing through the blackbox method. System testing was conducted among adolescent girls in the Sumber Sekar area and showed that all main features of the system functioned properly, with a functional accuracy rate of approximately 95% in blackbox testing. The system is also considered user-friendly and capable of providing early diagnostic insights and recommendations efficiently and accurately. It is expected to serve as an initial support tool to raise awareness and facilitate early detection of anemia, particularly among adolescent girls.

Implementation of Extract, Transform, Load (ETL) Method for Sales Data Visualization Using Google Looker Studio

Aliza Fatha Amanda — 2025-10-31

Café Durian Woke is a culinary business that focuses on selling various foods and beverages made from durian. The problem faced is that sales reports on the Olsera cashier system are still presented in tabular form, making it difficult for management to understand sales trends and make strategic evaluations and decisions. This study aims to develop sales data visualization for Café Durian Woke using Google Looker Studio tools with an Extract, Transform, Load (ETL) approach. The ETL method is used to extract transaction data, clean and transform the data using Microsoft Excel, and then load the results into Looker Studio for interactive visualization. The results of the study show that the developed dashboard is capable of displaying important information such as product sales trends, customer categories, and purchasing methods, where dine-in transactions dominate 93.3% of total sales. This visualization helps management identify superior products, understand customer behavior, and support faster, data-driven decision making.

Algoritma Transparent Data Encryption : Analisis dan Evaluasi Kinerja pada Sistem Keamanan Data At Rest SQL Server

Bita Parga Zen — 2025-10-31

This study evaluates the data-at-rest security enhancement and performance impact of Transparent Data Encryption (TDE) implementation on Microsoft SQL Server by comparing three conditions: no TDE, TDE AES-128, and TDE AES-256. Three databases with identical schemas were prepared. They were filled with a mix of text, numeric, image, and video data. The databases were then tested sequentially according to the methodology. The process included schema and integrity validation, SMK key hierarchy configuration, certificates, DE, encryption status verification, encryption and decryption throughput measurement, backup and recovery time evaluation, I/O stall analysis, and CPU and memory usage monitoring. Data-at-rest security validation was performed using file and log access tests, as well as cross-server recovery requiring keys. The results showed that TDE successfully eliminates file and log readability during direct access. It also requires the presence of keys during recovery, thus significantly reducing the risk of data exposure. Performance impacts are at acceptable levels: average CPU load increases around 1–2% and memory load increases by 1–8% under load. Encryption and decryption throughput between AES-128 and AES-256 are relatively equivalent. Backup operations tend to be slightly slower on TDE, while restores are faster, in line with lower read I/O stalls on the encrypted basis. Overall, TDE provides significant security enhancements with minimal performance compromise. Limitations of this study lie in the test load profile and hardware

Village Website Design and Implementation Using Laravel for Village Administration Services

Ahmad mujib — 2025-10-31

Administrative and information services in Simpang Tais Village were previously managed manually, requiring residents to visit the office for document submissions such as domicile letters and business permits. This conventional process often caused queues, delays, and limited transparency. Information dissemination was also restricted to bulletin boards or oral communication, reducing its reach to citizens. To overcome these challenges, this study designed and implemented a village website using the Laravel framework and Agile methodology. The development process included planning, implementation, testing, deployment, and maintenance stages. The website provides key features such as online letter submission, verification by the village secretary, approval by the village head, and content management for news and announcements. Black-box testing confirmed that all modules functioned properly according to user needs. The results show that the system improved efficiency by reducing service time, enhanced transparency through online submission tracking, and increased accessibility to information. In conclusion, the Laravel-based website developed with the Agile approach effectively supports digital transformation at the village level and provides a foundation for future integration with broader e-government systems

Prediksi Angka Harapan Hidup di Indonesia Berdasarkan Jenis Kelamin Menggunakan Regresi Linear

Khoirun Nisa — 2025-10-31

This study discusses the prediction of Life Expectancy (LE) in Indonesia based on gender for the period 2025 to 2029. LE is used as an indicator of societal welfare and quality of life. Historical data on male and female LE were obtained from the Central Bureau of Statistics and analyzed using the linear regression method. This research aims to identify the growth patterns of LE separately by gender. The regression model was developed by segregating the data based on gender, followed by model training, accuracy evaluation using the coefficient of determination, and analysis of the prediction results. This method produces a predictive model that can be utilized in the formulation of medium-term health policies. The study contributes to the development of a simple statistical model for LE projection specifically by gender, which has been rarely addressed in previous studies.

Implementasi Aplikasi Android Deteksi Penyakit Tanaman Tomat Berbasis Cloud menggunakan MVVM

M. Edoazani — 2025-10-31

Tomato plants as important horticultural commodities in Indonesia are vulnerable to various diseases that are difficult to distinguish visually, causing treatment delays and economic losses for farmers. This research develops an Android application for cloud-based tomato plant disease detection using Model-View-ViewModel (MVVM) architecture with Jetpack Compose framework. Extreme Programming (XP) methodology is applied through four main stages: planning, design, coding, and testing iteratively to ensure responsive development toward system requirements. The system integrates MobileNetV2 machine learning model deployed on Azure App Service with Firebase Authentication and Firestore for authentication and detection history storage. Dataset containing 16,012 tomato leaf images with 10 disease classes is used to train the model achieving 91.74% accuracy. Cloud-hybrid architecture enables detection processes to be performed on server without burdening mobile devices with average response time of 3 seconds. Black Box Testing shows all application features function according to specifications with 100% success rate. Research results prove that cloud computing integration with mobile application development can provide practical and scalable solutions to help farmers perform tomato plant disease detection quickly and accurately

Klasifikasi Data Kelulusan Mahasiswa Menggunakan Algoritma C.45

azwar anas — 2025-10-31

Perkembangan teknologi informasi dan ilmu komputer telah mendorong berbagai institusi, termasuk perguruan tinggi, untuk mengoptimalkan pemanfaatan data dalam proses pengambilan keputusan. Setiap mahasiswa memiliki himpunan data yang akan berbeda. Hal ini tergantung dari mahasiswa dalam kemampuan akademik maupun non akademik. Semakin besar jumlah mahasiswa, semakin beragam pula variasi nilai yang dihasilkan. Kumpulan data tersebut akan tampak seperti sekadar tumpukan informasi tanpa makna apabila tidak dilakukan analisis secara mendalam. Namun, melalui penerapan metode data mining, data tersebut dapat diolah menjadi pengetahuan baru yang sebelumnya tersembunyi dan tidak teridentifikasi secara eksplisit. Tujuan dari penelitian ini adalah untuk mengklasifikasin data mahasiswa berdasarkan dua target yaitu lulus tepat waktu dan tidak tepat waktu. Metode yang digunakan adalah algoritma klasifikasi C4.5 dan software data mining Weka. Data training yang digunakan berjulah 83. Hasil penelitian menunjukkan tingkat akurasi data diklasifikan dengan benar mencapai 91,5663% dan 61 mahasiswa akan lulus tepat waktu dari total 83 orang yang diklasifikasikan. Variabel dengan gain tertinggi adalah IPK dan diikuti oleh Beasiswa.

Perancangan Desain UI/UX Sistem Akademik Di SMKN 8 Palembang Menggunakan Metode Lean UX

Wardha Suada — 2025-10-31

The design of the academic information system at SMK Negeri 8 Palembang utilizes the Lean UX method to optimize user interface (UI) and user experience (UX). This study aims to solve inefficiencies in academic data management, prevent duplication, and reduce input errors commonly found in manual systems. The research process involved assumption declaration, development of a Minimum Viable Product (MVP) via Figma, experimentation using Maze, and usability evaluation through the System Usability Scale (SUS). Usability tests showed high completion rates, and the SUS score averaged 80.3, indicating excellent system acceptability. The results demonstrate that the developed UI/UX design offers an intuitive, efficient, and user-centered experience for students, teachers, and staff. This research contributes to digital transformation efforts in vocational education, providing a usable reference for the future implementation of web-based academic information systems.

Keywords: lean ux, academic information system, ui/ux design, usability testing, vocational school

Analisis Adopsi Pemanfaatan Teknologi Digital Pada Usaha Mikro Kecil Menggunakan Framework TOE

Riska Rahma Sari — 2025-10-31

The advancement of digital technology has a great influence on the development of Micro and Small
Enterprises (MSEs). However, the level of digital technology adoption by MSE actors is still relatively
low. This study aims to analyze the factors that influence the adoption of digital technology using the
Technology-Organization-Environment (TOE) framework. A quantitative approach was used with the
Partial Least Square Structural Equation Modeling (PLS-SEM) method through SmartPLS 4.0. The
research sample consisted of 97 MSEs in Jambi City engaged in the food and beverage sector. The
questionnaire instrument was prepared based on the three dimensions of TOE: Technology,
Organization, and Environment. The results showed that technological and environmental factors had a
significant effect on the adoption of digital technology, while organizational factors did not show a
significant effect. This may be due to the simple and informal organizational structure of MSEs. This
finding shows that the TOE framework is relevant in analyzing the adoption of digital technology in
MSEs and provides implications for business actors and policy makers to further emphasize technological
and environmental aspects in the digitalization strategy of MSEs.

Selection of the Best Teacher at SMP PGRI 4 Jambi City Using the TOPSIS and WP Methods

Firmansyah — 2025-10-31

Decision Support System is part of an information system designed to assist the decision-making process in complex and unstructured conditions. DSS can be used in the process of selecting the best teachers, one of which is at SMP PGRI 4 Kota Jambi. The main problem is that the assessment process is still carried out manually, causing delays in processing scores and potentially leading to information leaks. To overcome this, a web-based DSS for selecting the best teachers was developed, implementing the TOPSIS and Weighted Product (WP) methods to determine the best teachers. Based on the calculation results, the TOPSIS method showed superior performance compared to WP in terms of sensitivity, which was 5.917% for TOPSIS and 5.085% for WP. However, in terms of ranking stability, the WP method is considered more consistent, where the test results show that Yuhelmi, S.Pd (A2) was predominantly selected as the best teacher at every stage of testing. Thus, the selection of the method can be adjusted to the reference used: TOPSIS excels in sensitivity of results, while WP is more stable in ranking. In addition, this system has been proven to accelerate the evaluation process and improve the efficiency of selecting the best teachers

Analisis Website MAN 2 Jambi Menggunakan Webqual 4.0 Dan Importance Performance Analysis (IPA)

Hani Prastiwi — 2025-10-31

Kualitas website sekolah berperan penting dalam meningkatkan efektivitas komunikasi dan penyampaian informasi antara sekolah, siswa, orang tua, serta masyarakat. Namun, website MAN 2 Kota Jambi masih memiliki permasalahan seperti tampilan yang kurang responsif, navigasi belum optimal, dan keterlambatan pembaruan informasi yang dapat menurunkan kepuasan pengguna. Penelitian ini bertujuan untuk menganalisis kualitas website MAN 2 Kota Jambi menggunakan metode WebQual 4.0 yang terdiri atas dimensi Usability, Information Quality, dan Service Interaction Quality, serta metode Importance Performance Analysis (IPA) untuk menentukan prioritas perbaikan. Data dikumpulkan melalui penyebaran kuesioner kepada 384 responden, dengan mayoritas berusia di bawah 20 tahun (70,05%) dan berstatus pelajar. Hasil uji validitas dan reliabilitas menunjukkan semua indikator dinyatakan valid dan reliabel (nilai Cronbach’s Alpha > 0,60). Uji parsial (t) dan simultan (F) memperlihatkan bahwa ketiga dimensi WebQual berpengaruh signifikan terhadap kepuasan pengguna dengan nilai R² sebesar 0,691 atau 69,1%. Hasil pemetaan IPA menunjukkan bahwa Information Quality menjadi prioritas utama untuk diperbaiki, sedangkan Service Interaction Quality memiliki kinerja dan kepentingan tinggi sehingga perlu dipertahankan. Hasil penelitian ini memberikan rekomendasi strategis bagi pengelola MAN 2 Kota Jambi dalam meningkatkan kualitas website agar lebih efektif memenuhi kebutuhan pengguna.

Implementasi Model BiLSTM-Attention untuk Prediksi Nilai IHSG Berdasarkan Data Historis OHLCV

Amirah Rizky Ramadhanti — 2025-10-31

The Composite Stock Price Index (IHSG) reflects the performance of the Indonesian capital market, but predicting it is challenging due to high volatility and the influence of various external factors. This study aims to develop and evaluate a deep learning-based predictive model using a Bidirectional Long Short-Term Memory (BiLSTM) architecture combined with an Attention Mechanism to predict the IHSG value based on historical numerical data (OHLCV). This method was chosen for its ability to recognize bidirectional sequential patterns and highlight the most relevant historical information in the prediction process. The research was conducted quantitatively using an experimental approach, and model evaluation was performed using regression metrics such as R², RMSE, MAE, and MAPE. The results obtained showed excellent predictive performance with an R² of 0.9485, MAPE of 0.63%, RMSE of 59.47, and MAE of 45.12. Additionally, attention weight analysis revealed that the model focuses more on the last two days within the prediction time window, indicating that recent information significantly influences IHSG movements. These findings suggest that the BiLSTM-Attention approach is effective in capturing stock market dynamics and has the potential to serve as a strategic tool for data-driven investment decision-making.

Analysis of User Satisfaction Levels on Service Quality in the Grab Application Using the E-Servequal and Kano Methods

Baiti Rahmah Baiti — 2025-10-31

Grab is a digital offers a wide range of services, including ride-hailing, food delivery, parcel delivery, and digital payment solutions. Its presence has become an integral part of urban life, including in Jambi City, by providing convenient access and fast services. This study aims to analyze the service quality of the Grab application using the E-SERVQUAL approach integrated with the Kano model to identify service attributes that influence user satisfaction. A quantitative research method was applied, with data collected through an online questionnaire distributed to Grab users in Jambi City. The data were analyzed to measure the gap between users’ expectations and perceptions across five E-SERVQUAL dimensions: reliability, responsiveness, assurance, empathy, and tangibles, as well as to classify service attributes based on the Kano model. The findings reveal that reliability and responsiveness have the greatest influence on user satisfaction, while several attributes fall under the must-be category, which must be fulfilled to prevent a decline in satisfaction. These results are expected to serve as a strategic reference for Grab’s management to continuously improve its service quality.

Optimasi Kecepatan dan Responsivitas Aplikasi Presensi Perkuliahan Berbasis Web melalui Arsitektur Single Page Application

Ali Sadikin — 2025-10-31

Web-based lecture attendance applications that use Multi-Page Application (MPA) architecture often face speed and responsiveness issues when the number of users increases. This study aims to optimize the attendance application at Universitas Dinamika Bangsa (UNAMA) by implementing Single Page Application (SPA) architecture to improve system access speed and responsiveness. Using the waterfall development method, the results of this study show that SPA implementation is significantly more efficient than Multi-page Application (MPA) architecture in several aspects, including the number of Requests (RC), Transferred File Size (TF), Resources (RS), and Request Duration (RD). For example, the “Home” page on MPA requires 51 requests with a total transfer of 741 kB and a duration of 945 ms, while on SPA it only requires 1 request with a transfer of 5.9 kB and a duration of 187 ms. This proves that the SPA architecture successfully reduces network load and speeds up response time, resulting in a more responsive user experience and efficient use of resources.

Pengembangan Aplikasi Monitoring Kondisi Tanaman Berbasis Markerless Augmented Reality dengan Metode MDLC

Muhammad Gemilang Ramadhan — 2025-10-31

Augmented Reality (AR) semakin banyak diterapkan dalam bidang pertanian untuk membantu proses monitoring dan visualisasi data lingkungan. Namun, sebagian besar aplikasi AR yang ada masih bergantung pada penggunaan marker dan belum banyak mengintegrasikan data sensor secara real-time. Penelitian ini mengembangkan aplikasi markerless AR untuk memantau kondisi tanaman dengan menerapkan metode Multimedia Development Life Cycle (MDLC). Tahapan pengembangan meliputi konseptualisasi, perancangan, pengumpulan material, perakitan, pengujian, dan distribusi. Aplikasi dibangun menggunakan Unity dan AR Foundation dengan pendekatan plane tracking untuk menampilkan informasi kondisi tanaman secara visual melalui teks dan angka digital. Data dari sensor lingkungan diintegrasikan melalui mikrokontroler ESP32, namun fokus penelitian ini adalah penyajian visual, bukan kontrol IoT. Hasil pengujian menunjukkan bahwa aplikasi mampu menampilkan data secara real-time dengan stabilitas visual yang baik pada kondisi pencahayaan optimal. Penelitian ini menunjukkan bahwa penerapan MDLC dapat mendukung pengembangan aplikasi markerless AR yang terstruktur, efisien, dan relevan untuk sistem monitoring pertanian.