Jurnal Sistem Informasi dan Ilmu Komputer https://jurnal.unprimdn.ac.id/index.php/JUSIKOM <p>Jurnal Sistem Informasi dan Ilmu Komputer Prima (JUSIKOM PRIMA) is published by the Information Systems study program, Faculty of Technology and Computer Science, University of Prima Indonesia (UNPRI) Medan as a medium for publishing scientific papers for researchers. The scientific works produced are in the form of qualitative and quantitative research results, information system design, analysis and application program design. This journal is published twice a year, in <strong>February</strong> and <strong>August</strong>.</p> <p>The Jurnal JUSIKOM already has<br />E - ISSN: <a title="E-ISSN" href="https://issn.brin.go.id/terbit/detail/1493265251" target="_blank" rel="noopener">2580-2879</a></p> <div class="style1" align="justify">The submitted paper will be reviewed by reviewers. Review process employs <strong>Double-Blind Peer Review.</strong>In this system authors do not know who the reviewer is, and the reviewers do not know whose work they are evaluating.</div> <div class="style1" align="justify">Before submission, please <strong>make sure that your paper </strong>is prepared using the journal <strong><a href="http://jurnal.unprimdn.ac.id/index.php/JUSIKOM/libraryFiles/downloadPublic/3" target="_blank" rel="noopener">Paper Template</a>. </strong></div> <div class="style1" align="justify"><strong><br />Online Submissions </strong></div> <div class="style1" align="justify">Already have a Username/Password for Jurnal Jusikom? <strong><a href="http://jurnal.unprimdn.ac.id/index.php/JUSIKOM/login" target="_blank" rel="noopener">GO TO LOGIN</a>. </strong></div> <div class="style1" align="justify">Need a username/password? <strong><a href="http://jurnal.unprimdn.ac.id/index.php/JUSIKOM/user/register?source=">GO TO REGISTRATION</a>. <br />Registration and login are required to submit items online and to check the status of current submissions. <br /></strong></div> <p> </p> Fakultas Teknologi dan Ilmu Komputer Universitas Prima Indonesia en-US Jurnal Sistem Informasi dan Ilmu Komputer 2580-2879 <p>Authors who publish their manuscripts through the Journal of Information Systems and Computer Science agree to the following:</p> <ul> <li>Copyright to the manuscripts of scientific papers in this Journal is held by the author.</li> <li>The author surrenders the rights when first publishing the manuscript of his scientific work and simultaneously the author grants permission / license by referring to the <strong>Creative Commons Attribution-ShareAlike 4.0 International License</strong> to other parties to distribute his scientific work while still giving credit to the author and the Journal of Information Systems and Computer Science as the first publication medium for the work.</li> <li>Matters relating to the non-exclusivity of the distribution of the Journal that publishes the author's scientific work can be agreed separately (for example: requests to place the work in the library of an institution or publish it as a book) with the author as one of the parties to the agreement and with credit to sJournal of Information Systems and Computer Science as the first publication medium for the work in question.</li> <li>Authors can and are expected to publish their work online (e.g. in a Repository or on their Organization's/Institution's website) before and during the manuscript submission process, as such efforts can increase citation exchange earlier and with a wider scope.</li> </ul> ANALISIS SENTIMEN ULASAN PRODUK MENGGUNAKAN LARGE LANGUAGE MODELS: STUDI KASUS PADA SHOPEE https://jurnal.unprimdn.ac.id/index.php/JUSIKOM/article/view/6601 <p>Perkembangan pesat <em>e-commerce</em> telah meningkatkan jumlah ulasan produk yang tersedia di platform seperti Shopee. Ulasan ini dapat menjadi sumber informasi berharga bagi penjual dalam memahami persepsi konsumen dan meningkatkan strategi pemasaran. Namun, besarnya volume dan kompleksitas bahasa dalam ulasan membuat analisis manual menjadi tidak efisien. Penelitian ini bertujuan untuk menganalisis sentimen ulasan produk di Shopee menggunakan Large Language Models (LLMs), khususnya model Gemini 1.5-Pro yang telah di-fine-tune agar lebih sesuai dengan bahasa pengguna Shopee. Metode yang digunakan mencakup pengumpulan data melalui web scraping, preprocessing data, fine-tuning model, serta evaluasi performa model berdasarkan metrik akurasi, presisi, recall, dan F1-score. Hasil penelitian menunjukkan bahwa model yang dikembangkan mampu mengklasifikasikan sentimen ulasan ke dalam kategori positif, negatif, dan netral dengan akurasi berkisar antara 66,67% hingga 85,71%.</p> Ribka Amelia Yunita Keliat Evta Indra Yonata Laia Copyright (c) 2025 Ribka Amelia Yunita Keliat, Evta Indra, Yonata Laia https://creativecommons.org/licenses/by-sa/4.0 2025-03-15 2025-03-15 8 2 Exploratory Data Analysis Historical Cryptocurrency https://jurnal.unprimdn.ac.id/index.php/JUSIKOM/article/view/6548 <p>Penelitian ini menganalisis data historis <em>cryptocurrency</em> menggunakan metode <em>Exploratory Data Analysis</em> (EDA) untuk membantu investor pemula memahami pergerakan harga mata uang kripto. <em>Cryptocurrency</em>, sebagai uang digital yang tidak berwujud fisik, memiliki tingkat volatilitas tinggi yang sering menyebabkan kerugian bagi investor yang kurang berpengalaman dalam menganalisis data historis. Menggunakan dataset dari <em>CoinMarketCap</em> yang terdiri dari 679.183 baris dan 13 kolom periode 2017-2022, penelitian ini menerapkan metodologi EDA dengan pendekatan visualisasi data. Prosedur penelitian mencakup analisis masalah, data <em>acquisition</em> melalui web <em>scraping</em>, data cleaning, dan <em>exploratory data analysis</em>.</p> <p>Hasil analisis menunjukkan persaingan antara Bitcoin dan Ethereum. Berdasarkan marketval, Bitcoin mencapai $140.000.000.000, sementara Ethereum $80.000.000.000. Volume Bitcoin mencapai $8.000.000.000, sedangkan Ethereum $4.000.000.000. Analisis <em>price movement</em> menunjukkan <em>Ethereum</em> mencapai $140.000, sementara Bitcoin $1. Dalam analisis <em>moving average</em>, <em>Ethereum</em> menunjukkan performa lebih baik dengan grafik mencapai $105.000, dibandingkan Bitcoin yang hanya mencapai $0,8. Penelitian ini berkontribusi dalam membantu investor pemula memahami dinamika pasar <em>cryptocurrency</em> melalui analisis data historis. Hasil visualisasi dan analisis dapat digunakan sebagai acuan pengambilan keputusan investasi dan meminimalisir risiko kerugian. Studi ini merekomendasikan penggunaan data historis sebagai alat prediksi dibandingkan pengambilan keputusan berbasis intuisi dalam investasi <em>cryptocurrency</em>.</p> Ezra Christina Septiana Panjaitan Evta Indra Copyright (c) 2025 Ezra Christina Septiana Panjaitan, Evta Indra https://creativecommons.org/licenses/by-sa/4.0 2025-02-27 2025-02-27 8 2 90 100 Analysis and Design of Web-Based Inventory Receipt and Management Information Systems at Heycaps.Co Stores Using the Prototype Method https://jurnal.unprimdn.ac.id/index.php/JUSIKOM/article/view/5965 <p>Effective inventory management is crucial for Heycaps.co to ensure smooth operations and customer satisfaction. However, the current manual system results in inventory data inconsistencies, delays in the receipt of goods, and challenges in inventory tracking. This study aims to design a web-based inventory receipt and management information system to address these issues. The system is developed using a prototype approach, incorporating Flowcharts, Data Flow Diagrams (DFD), Entity Relationship Diagrams (ERD), Table Relations, and Database Table Structures. The interface design employs the Bootstrap framework to ensure a responsive and user-friendly display. The findings of this study present a system design that can be utilized by the <em>owner</em>, warehouse staff, and store staff to enhance inventory data management, streamline the receipt process, and improve the accuracy of inventory reporting. User evaluations indicate that the proposed system meets user requirements and offers ease of use.</p> <p><strong>Keywords</strong>: Information System, Inventory, Prototype Method, Heycaps.co, Inventory Management.</p> Ni Putu Tia Ananda I Gusti Agung Pramesti Dwi Putri Ni Putu Noviyanti Kusuma Copyright (c) 2025 Ni Putu Tia Ananda, I Gusti Agung Pramesti Dwi Putri, Ni Putu Noviyanti Kusuma https://creativecommons.org/licenses/by-sa/4.0 2025-02-28 2025-02-28 8 2 75 89 10.34012/jurnalsisteminformasidanilmukomputer.v8i2.5965 Current Trends and Future Directions of Big Data in Commerce: A Bibliometric Analysis Based on Scopus https://jurnal.unprimdn.ac.id/index.php/JUSIKOM/article/view/6098 <p>Big data provides significant benefits across various sectors, including commerce. However, there remained a gap in bibliometric studies examining big data within the context of commerce, leaving research development in this field unclear. This study aimed to address this gap by conducting a bibliometric investigation into researchers' contributions to big data in commerce, including their affiliations and countries of origin. Additionally, the study sought to identify the most productive journals and highlight relevant and under-researched topics within this field. A bibliometric analysis approach was employed, analyzing 396 Scopus-indexed documents and using VOSviewer visualization to identify major recurring issues in the literature. The findings revealed that in 2021, the number of publications on big data in commerce peaked at 97 documents. Maalla, A., from Guangzhou College of Technology and Business, China, emerged as the most prolific author, while China led in publication output with 308 documents. The Journal of Physics Conference Series was identified as the most productive source. Computer Science was the most explored discipline, indicating a strong integration of technology with commerce. Keyword analysis divided research focus into four main clusters: analytical technology, platform optimization, supply chain management, and marketing strategy optimization. These findings provide a foundation for future research to explore areas such as Customer Experience Management, Blockchain Technology, Cloud Computing, Predictive Analytics, and Customer Segmentation, thereby enriching the academic literature and offering practical contributions to data-driven commerce.</p> Ahmad Bilal Almagribi Bambang Purnomosidi Dwi Putranto Copyright (c) 2025 Ahmad Bilal Almagribi, Bambang Purnomosidi Dwi Putranto https://creativecommons.org/licenses/by-sa/4.0 2025-02-28 2025-02-28 8 2 57 74 10.34012/jurnalsisteminformasidanilmukomputer.v8i2.6098 The Impact of Incremental Innovation at Gojek Startup on Users in Batam City Using the Expectation Confirmation Model https://jurnal.unprimdn.ac.id/index.php/JUSIKOM/article/view/6205 <h3>In this modern era, the world has witnessed a phenomenal explosion in the growth of startups in various countries. Startups have become the main drivers of innovation and economic growth. Startup businesses are currently experiencing significant and rapid growth, especially in Indonesia. Currently, there are many new startup companies, so innovation is needed to compete with other competitors. Startup development requires creative ideas to attract investors. In addition to innovative ideas, product quality must also be a priority to attract consumer interest. Therefore, this study aims to examine the impact of incremental innovation on startups from the perspective of users in Batam City. The purpose of this study is to examine the impact of incremental innovation on the Gojek application. The author uses the Structural Equation Model (SEM) with the Partial Least Squares method, then the researcher applies the Extended Expectation Confirmation Model (ECM) from the perspective of Gojek application users to analyze the effect of incremental innovation efforts on the Gojek application. The author collects data by distributing questionnaires to people who have made transactions with Gojek. A total of 264 samples have been collected, and the results show that confirmation has a positive effect on the perception of enjoyment, satisfaction, and customer engagement. The results show that the perception of enjoyment and satisfaction has a positive impact on the user's intention to continue using the Gojek application. However, customer engagement does not significantly affect the user's intention to continue using the Gojek application.</h3> Vincent Vincent Indasari Deu Eryc Eryc Copyright (c) 2025 Vincent Vincent, Indasari Deu, Eryc Eryc https://creativecommons.org/licenses/by-sa/4.0 2025-02-21 2025-02-21 8 2 47 56 10.34012/jurnalsisteminformasidanilmukomputer.v8i2.6205 Core Banking Testing Pada Fitur Customer Transaction di BPR Lestari Bali https://jurnal.unprimdn.ac.id/index.php/JUSIKOM/article/view/6411 <p>BPR Lestari Bali adalah salah satu Bank Perkreditan Rakyat (BPR) terbesar di Bali yang dikenal sebagai pelopor dalam memberikan layanan keuangan kepada masyarakat lokal. Saat ini, BPR Lestari tengah merumuskan strategi untuk meningkatkan kualitas layanan dan efisiensi aktivitas perbankan melalui implementasi sistem terintegrasi, yaitu <em>Core Banking System</em> (CBS). CBS merupakan platform yang dirancang untuk mengelola berbagai operasi inti perbankan, seperti penyimpanan data nasabah, transaksi harian, dan layanan digital. Penelitian ini dilakukan dengan tujuan untuk menguji CBS yang telah dikembangkan menggunakan metode <em>BlackBox Testing</em> dengan teknik <em>Equivalence Partitioning</em>, yang menekankan evaluasi fungsionalitas sistem tanpa menganalisis struktur internal kode. Pengujian dilakukan fokus pada fitur <em>Customer Transaction</em>, yang terdiri dari enam modul utama, yaitu: <em>Transaction, Deposit, Customer, Collateral, Account</em> &amp; <em>Wallet</em>, serta <em>Lending</em>. Hasil pengujian menunjukkan tingkat keberhasilan mencapai 98,38%. Lima dari enam modul berhasil menunjukkan tingkat keberhasilan sebesar 100%, sedangkan modul <em>Customer</em> mencapai tingkat keberhasilan 87,5% akibat adanya <em>bug</em> yang memerlukan penyelesaian lebih lanjut oleh <em>developer</em>. Dari pengujian ini, dapat disimpulkan bahwa metode <em>BlackBox Testing</em> dengan teknik <em>Equivalence Partitioning</em> efektif dalam memastikan fungsionalitas CBS, mengurangi risiko gangguan layanan, dan meningkatkan kepercayaan pengguna terhadap sistem.</p> Putu Arya Novianingsih Ni Putu Yuliawati Nengah Widya Utami I Gusti Agung Pramesti Dwi Putri Copyright (c) 2025 Putu Arya Novianingsih Ni Putu Yuliawati, Nengah Widya Utami, I Gusti Agung Pramesti Dwi Putri https://creativecommons.org/licenses/by-sa/4.0 2025-02-26 2025-02-26 8 2 36 46 Analisis dan Perancangan Sistem Keuangan Universitas Primakara Menggunakan Unified Modeling Language (UML) dengan Metode Agile https://jurnal.unprimdn.ac.id/index.php/JUSIKOM/article/view/6415 <h3>Perkembangan teknologi informasi mendorong efisiensi dalam pengelolaan informasi, termasuk implementasi sistem informasi. Universitas Primakara membutuhkan sistem informasi keuangan untuk mengatasi pengelolaan tagihan UKT mahasiswa yang masih manual dan kurang terintegrasi. Penelitian ini bertujuan merancang pemodelan sistem keuangan menggunakan Unified Modeling Language (UML) dengan metode Agile. Pemodelan dilakukan melalui use case diagram, activity diagram, sequence diagram, dan desain database menggunakan Entity Relationship Diagram (ERD). Hasil penelitian menunjukkan bahwa metode Agile dengan kerangka kerja Scrum efektif dalam menghasilkan pemodelan sistem yang sesuai dengan kebutuhan pengguna. Pemodelan ini mempermudah pengelolaan tagihan UKT secara terintegrasi. Kesimpulannya, metode Agile dan UML terbukti adaptif dan komprehensif untuk perancangan sistem informasi keuangan Universitas Primakara.</h3> Hananindita Djohan I Gusti Agung Pramesti Dwi Putri I Putu Buda Suyasa Copyright (c) 2025 Hananindita Djohan, I Gusti Agung Pramesti Dwi Putri, I Putu Buda Suyasa https://creativecommons.org/licenses/by-sa/4.0 2025-02-27 2025-02-27 8 2 12 35 Comparative Analysis of the Customer Satisfaction Index and Service Quality Methods in Measuring BPJS Patient Satisfaction at Royal Prima Hospital https://jurnal.unprimdn.ac.id/index.php/JUSIKOM/article/view/6508 <p>Patient satisfaction is an important indicator for assessing the quality of healthcare services, especially for hospitals serving BPJS patients. By understanding the level of patient satisfaction, hospitals can identify what is performing well and what needs improvement in their services. The objective of this study is to compare two methods, namely the Customer Satisfaction Index (CSI) and Service Quality (Servqual), in measuring BPJS patient satisfaction at Royal Prima Hospital. The CSI method quantitatively measures overall patient satisfaction, while the Servqual method evaluates satisfaction based on four service quality dimensions. This research was conducted through questionnaires distributed to 300 BPJS patients who had received medical services at Royal Prima Hospital. The research findings indicate that the Servqual method produced an average satisfaction score of 2.74, while the CSI method achieved a satisfaction score of 0.69, which falls into the "Good" or "Satisfied" category. These findings demonstrate that both methods complement each other, providing a more effective and comprehensive understanding of patient satisfaction.</p> Ria Putri Zevanya Christine Ester Novita Sari Rafael Crisman MPH Muhammad Rusdi Batubara Donni Nasution Copyright (c) 2025 Ria Putri Zevanya, Christine Ester Novita Sari , Rafael Crisman MPH, Muhammad Rusdi Batubara, Donni Nasution https://creativecommons.org/licenses/by-sa/4.0 2025-02-21 2025-02-21 8 2 01 11 Analisis Akurasi Algoritma K-Nearest Neighbor Untuk Diagnosis Penyakit Jantung Pada Lansia https://jurnal.unprimdn.ac.id/index.php/JUSIKOM/article/view/6538 <p>Penyakit jantung adalah salah satu penyebab utama kematian di dunia, terutama pada populasi lansia, yang sering kali sulit dideteksi pada tahap awal karena gejala yang tidak spesifik. Oleh karena itu, diperlukan metode diagnosis yang lebih cepat dan efisien, seperti penerapan algoritma pembelajaran mesin. Penelitian ini bertujuan untuk mengevaluasi akurasi algoritma K-Nearest Neighbor (KNN) dalam mendiagnosis penyakit jantung pada lansia, dengan menggunakan dataset yang diperoleh dari Kaggle dan terdiri dari 918 data pasien. Data tersebut disaring untuk usia lansia (60 tahun ke atas), menghasilkan 253 data yang digunakan dalam klasifikasi. Empat nilai k (3, 5, 7, dan 9) diuji untuk menentukan nilai k terbaik dalam mengklasifikasikan penyakit jantung. Hasil evaluasi menunjukkan bahwa model dengan k = 9 memiliki performa terbaik dengan nilai recall tertinggi (0.93) dan F1-Score sebesar 0.81, meskipun dengan akurasi yang sedikit lebih rendah (0.68). K = 5 memberikan keseimbangan terbaik antara precision (0.72) dan recall (0.85), dengan F1-Score 0.78. Berdasarkan hasil ini, K = 9 lebih efektif untuk aplikasi medis yang mengutamakan deteksi lebih banyak kasus positif, meskipun mengorbankan sedikit precision. Penelitian ini dapat memberikan kontribusi untuk pengembangan sistem diagnosis penyakit jantung yang lebih cepat, efisien, dan akurat pada lansia, dengan harapan dapat meningkatkan deteksi dini penyakit jantung.</p> David Sebastian Sipayung Syarifah Atika Copyright (c) 2025 David Sebastian Sipayung, Syarifah Atika https://creativecommons.org/licenses/by-sa/4.0 2025-03-06 2025-03-06 8 2