cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Bahar
Contact Email
bahararahman@gmail.com
Phone
-
Journal Mail Official
puslit.stmikbjb@gmail.com
Editorial Address
-
Location
Kota banjarmasin,
Kalimantan selatan
INDONESIA
Jutisi: Jurnal Ilmiah Teknik Informatika dan Sistem Informasi
Published by Sekolah Tinggi Manajemen Informatika dan Komputer Banjarbaru
ISSN : 20893787     EISSN : 26850893     DOI : -
Core Subject : Science,
Computer Science & IT
Jutisi: Jurnal Ilmiah Teknik Informatika dan Sistem Informasi adalah Jurnal Ilmiah bidang Teknik Informatika dan Sistem Informasi yang diterbitkan secara periodik tiga nomor dalam satu tahun, yaitu pada bulan April, Agustus dan Desember. Redaksi Jutisi: Jurnal Ilmiah Teknik Informatika dan Sistem Informasi menerima sumbangan tulisan hasil penelitian atau atau artikel konseptual bidang Teknik Informatika dan Sistem Informasi.
Arjuna Subject : -
Articles 1,023 Documents
 99   100   101   102   103 
Implementasi Sistem Lost and Found Kampus Berbasis Web Terintegrasi Geolocation dan Evaluasi MOS Zickrian, Firdaus Khotibul; Kurnia, Maulida Cahya; Devyana, Amanda; Apriyanto, Andika; Ramadhani, Ryandika Syauqi; Pramita, Cinantya
Jutisi : Jurnal Ilmiah Teknik Informatika dan Sistem Informasi Vol 15, No 1 (2026): Februari 2026
Publisher : STMIK Banjarbaru

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35889/jutisi.v15i1.3476

Abstract

This study developed a web-based Lost and Found system to address the issue of reporting lost items on campus, which was previously done through informal and decentralized media. The system was designed using a User-Centered Design approach and equipped with features for reporting, searching, category filtering, photo uploading, and geolocation-based location tagging. Evaluation was conducted through black box testing and user acceptance testing using Mean Opinion Score (MOS). The test results showed that all functions worked according to specifications, while the average MOS value of 4.98 indicated a very good level of user acceptance. These findings confirm that the developed system is capable of improving the effectiveness of the item retrieval process and is suitable for use as a service to support academic activities.Keywords: Lost and found; Geolocation; User-centered design; Mean opinion score; Information systemAbstrakPenelitian ini mengembangkan sistem Lost and Found berbasis web untuk mengatasi permasalahan pelaporan barang hilang di kampus yang sebelumnya dilakukan melalui media informal dan tidak terpusat. Sistem dirancang menggunakan pendekatan User-Centered Design dan dilengkapi fitur pelaporan, pencarian, filter kategori, unggah foto, serta penandaan lokasi berbasis geolocation. Evaluasi dilakukan melalui pengujian black box dan pengujian penerimaan pengguna menggunakan Mean Opinion Score (MOS). Hasil pengujian menunjukkan seluruh fungsi berjalan sesuai spesifikasi, sedangkan nilai MOS rata-rata sebesar 4,98 mengindikasikan tingkat penerimaan pengguna yang sangat baik. Temuan ini menegaskan bahwa sistem yang dikembangkan mampu meningkatkan efektivitas proses temu-kembali barang dan layak digunakan sebagai layanan pendukung aktivitas sivitas akademika. 
Utilization of the Google Spreadsheet Platform in the Development of a Marker Reporting System MAULANA, IVAN K RISQI; WISMARINI, T.D.
Jutisi : Jurnal Ilmiah Teknik Informatika dan Sistem Informasi Vol 15, No 1 (2026): Februari 2026
Publisher : STMIK Banjarbaru

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35889/jutisi.v15i1.3503

Abstract

Manual marker reporting at PT Anugerah Guna Abadi causes delays and data redundancy, hindering operational efficiency. This study aims to implement a desktop-based marker reporting application integrated with Google Spreadsheet to improve data efficiency, accuracy, and collaboration. The Research and Development (R&D) method with the Waterfall model was used to develop the system using VB.NET and the Google Sheets API. Testing covered functionality (blackbox), usability (SUS), and performance comparison with the manual system. The results showed the system successfully reduced processing time by 88.3% (from 18 to 2.1 minutes/transaction), decreased data error rates by 96.4% (from 18.67% to 0.67%), and enabled real-time collaboration with an SUS score of 84.7 (excellent). This research introduces a novel hybrid desktop-cloud architecture for marker reporting in the garment industry.Keywords: Marker reporting; Desktop application; Data synchronization; Real-time collaboration; garment industryAbstrakPelaporan marker manual di PT Anugerah Guna Abadi menyebabkan keterlambatan dan redundansi data, sehingga menghambat efisiensi operasional. Penelitian ini bertujuan mengimplementasikan aplikasi laporan marker berbasis desktop terintegrasi Google Spreadsheet untuk meningkatkan efisiensi, akurasi, dan kolaborasi data. Metode Research and Development (R&D) dengan model Waterfall digunakan untuk pengembangan sistem menggunakan VB.NET dan Google Sheets API. Pengujian mencakup fungsionalitas (blackbox), usabilitas (SUS), dan perbandingan kinerja dengan sistem manual. Hasil penelitian menunjukkan sistem berhasil mengurangi waktu proses sebesar 88,3% (dari 18 menjadi 2,1 menit/transaksi), menurunkan tingkat kesalahan data sebesar 96,4% (dari 18,67% menjadi 0,67%), dan memungkinkan kolaborasi real-time dengan skor SUS 84,7 (excellent). Penelitian ini menghasilkan kebaruan berupa arsitektur hybrid desktop-cloud untuk pelaporan marker di industri garmen. 
Classification of Chili Plant Pests Using the ConvNeXt Architecture Jocelyn, Jennifer; Devella, Siska
Jutisi : Jurnal Ilmiah Teknik Informatika dan Sistem Informasi Vol 15, No 1 (2026): Februari 2026
Publisher : STMIK Banjarbaru

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35889/jutisi.v15i1.3461

Abstract

Chili (Capsicum annuum L.) is a high-value horticultural commodity in Indonesia; however, its productivity often declines due to pest attacks that cause significant economic losses. This study aims to compare the performance of several ConvNeXt variants (V1 and V2) for chili pest classification using the Red Chili Pepper Pest dataset, which consists of four pest classes annotated with bounding boxes. The data were divided into training and testing sets, and a cropping process was applied to the object regions to ensure that the model focuses on pest images. The preprocessing stages included resizing, normalization, and data augmentation to improve model robustness against variations in image conditions. Model training was conducted using the timm library with uniform hyperparameter settings across all variants to ensure a fair comparison. Performance evaluation was carried out using accuracy, precision, recall, F1-score, and Area Under the Curve (AUC). In addition, computational complexity was analyzed in terms of the number of parameters, FLOPs, and inference latency. The results indicate that ConvNeXt V2 variants, particularly Nano and Tiny, achieve very high classification performance (macro-AUC > 0.99) with fewer parameters and lower latency compared to larger models. Robustness evaluation under various image degradations shows that Gaussian noise has the most significant negative impact on performance. Overall, ConvNeXt V2-Nano and V2-Tiny are recommended as the most efficient and stable models for implementing chili pest detection systems on resource-constrained devices within precision agriculture applications.Keywords: Chili Pest Classification; ConvNeXt; Deep Learning; Image Processing; Smart Agriculture.AbstrakCabai (Capsicum annuum L.) merupakan komoditas hortikultura bernilai tinggi di Indonesia, namun produktivitasnya sering menurun akibat serangan hama yang menyebabkan kerugian ekonomi. Penelitian ini bertujuan membandingkan kinerja varian ConvNeXt (V1 dan V2) dalam klasifikasi hama cabai menggunakan dataset Red Chili Pepper Pest yang terdiri atas empat kelas hama dengan anotasi bounding box. Data dibagi menjadi data pelatihan dan pengujian, kemudian dilakukan proses cropping pada objek untuk memastikan model berfokus pada citra hama. Tahapan prapemrosesan meliputi resizing, normalisasi, dan augmentasi untuk meningkatkan ketahanan model terhadap variasi citra. Pelatihan model dilakukan menggunakan pustaka timm dengan pengaturan hiperparameter pada seluruh varian untuk menjamin perbandingan adil. Evaluasi dilakukan menggunakan akurasi, presisi, recall, F1-score, dan AUC, serta analisis kompleksitas melalui jumlah parameter, FLOPs, dan latensi inferensi. Hasil penelitian menunjukkan ConvNeXt V2, khususnya Nano dan Tiny, mencapai performa tinggi (macro-AUC > 0,99) dengan kompleksitas komputasi lebih rendah. Uji robustness menunjukkan Gaussian noise memberikan penurunan performa paling signifikan. 
Analisis Sentimen Ulasan Pengguna Aplikasi M-Pajak di Google Play Store Menggunakan Algoritma Naïve Bayes Azahra, Nindita Nashwa; Arochman, Arochman; Ismanto, Bambang
Jutisi : Jurnal Ilmiah Teknik Informatika dan Sistem Informasi Vol 15, No 1 (2026): Februari 2026
Publisher : STMIK Banjarbaru

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35889/jutisi.v15i1.3402

Abstract

Digital-based public services require service quality evaluation to ensure user satisfaction, one way of doing this is through sentiment analysis of M-tax app reviews on the Google Play Store. The aim of this study is to test the tendency of user sentiment toward the M-Tax application by applying the naïve bayes algorithm. The dataset utilized in this research consists of 5,263 user reviews, which were processed through several preprocessing steps, including case folding, text cleaning, tokenization, stopword removal, and stemming. The analyzed variable is user review text classified into positive and negative sentiment categories. The naïve bayes algorithm was employed to perform sentiment classification. To assess the model performance, a confusion matrix was employed with an 80:20 split between train and testt data, along with k-fold cross-valiidation using k = 5. The findings show that negative sentiment dominates the reviews at 88.34%, while positive sentiment accounts for 11.66%. The model attained an accuracy of 92%, indicating that the Naïve Bayes algorithm performs efecctively in classifying user sentiment and can serve as a basis for evaluating improvements in digital taxation services.Keywords: Sentiment analysis; Naive Bayes; Application reviews; TF-IDF; M-Pajak AbstrakPelayanan publik berbasis digital menuntut adanya evaluasi kualitas layanan untuk memastikan kepuasan pengguna, salah satunya melalui analis sentiment ulasan aplikasi M-Pajak di Google Play Store. Adapun tujuan dari penelitian ini yaitu untuk mengkaji kecenderungan sentimen pada ulasan pengguna aplikasi M-Pajak di Google Play Store dengan menerapkan algoritma Naïve Bayes. Data yang dianalisis berjumlah 5.263 ulasan pengguna, yang diproses dengan tahapan prerocessing yaitu case folding, cleaning text, tokeniation, stopword, dan stemming. Variabel yang dianalisis berupa teks ulasan pengguna dengan kelas sentimen positif dan negatif. Klasifikasi dilakukan dengan menggunakan algoritma Naïve Bayes. Untuk menilai performa model, digunakan confusiont matrix sebagai alat evaluasi dengan pembagian data latih dan data sebesar 80:20 serta teknik k-fold cross-validation dengan nilai k = 5. Penelitian ini menunjukkan hasil bahwa sentimen negatif mendominasi sebesar 88,34%, sedangkan sentimen positif sebesar 11,66%. Model menghasilkan akurasi evaluasi sebesar 92%, sehingga algoritma Naïve Bayes dinilai tepat dan efisien untuk menganalisis sentimen pada ulasan aplikasi M-Pajak. dan dapat digunakan sebagai dasar evaluasi peningkatan layanan digital perpajakan. 
Sistem Informasi Cuti Pegawai Pada Badan Kepegawaian dan Pengembangan Sumber Daya Manusia Kabupaten Lampung Utara Siska, Siska; Hartono, Hartono
Jutisi : Jurnal Ilmiah Teknik Informatika dan Sistem Informasi Vol 15, No 1 (2026): Februari 2026
Publisher : STMIK Banjarbaru

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35889/jutisi.v15i1.3392

Abstract

Leave management at the BKPSDM of North Lampung Regency currently relies on manual procedures, risking recording errors and slow monitoring of leave balances. This research aims to design and implement an integrated digital employee leave information system. The system development follows the Waterfall model, spanning from analysis to implementation. Data collection involved workflow observations, staff interviews, and literature reviews. System testing was conducted using the Black Box Testing method to ensure all features function as planned. The results demonstrate that the system successfully automates leave requests and approvals transparently. In conclusion, this system minimizes data errors, provides real-time access to leave balance information, and enhances the efficiency and professionalism of personnel administration at BKPSDM North Lampung.Keywords: Information System; Leave; BKPSDM; Waterfall  AbstrakPengelolaan cuti pada BKPSDM Kabupaten Lampung Utara saat ini masih bergantung pada prosedur manual, yang berisiko menimbulkan kesalahan pencatatan dan lambatnya pemantauan saldo cuti. Penelitian ini bertujuan untuk merancang dan mengimplementasikan sistem informasi cuti pegawai digital yang terintegrasi. Pengembangan sistem mengikuti model Waterfall, mulai dari tahap analisis hingga implementasi. Pengumpulan data dilakukan melalui observasi alur kerja, wawancara staf, dan studi pustaka. Pengujian sistem dilakukan menggunakan metode Black Box Testing untuk memastikan seluruh fitur berfungsi sesuai rencana. Hasil penelitian menunjukkan bahwa sistem berhasil mengotomatisasi pengajuan dan persetujuan cuti secara transparan. Kesimpulannya, sistem ini meminimalkan kesalahan data, menyediakan akses informasi saldo cuti secara real-time, serta meningkatkan efisiensi dan profesionalisme administrasi kepegawaian di BKPSDM Lampung Utara. 
Implementasi Laravel dan AES-256 untuk Keamanan Arsip Digital PMB Oktoreza, Yunita Adelia; Wismarini, Th. Dwiati
Jutisi : Jurnal Ilmiah Teknik Informatika dan Sistem Informasi Vol 15, No 1 (2026): Februari 2026
Publisher : STMIK Banjarbaru

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35889/jutisi.v15i1.3511

Abstract

Operational obstacles in document retrieval and significant data security risks within the STIE BPD Jateng PMB Unit served as the primary motivation for this study. The researcher proposed a web-based archiving platform developed using the Laravel framework, reinforced by the Advanced Encryption Standard (AES-256) encryption scheme. The development procedure followed the Waterfall methodology, encompassing requirements identification, structural design, and the coding phase. Research findings indicated that utilizing the MVC pattern in Laravel produced a more organized system, while the integration of the AES-256 algorithm effectively protected documents from unauthorized access. Based on black-box testing results, all functional components of the system were proven to operate optimally. It was concluded that the implementation of this technology not only significantly accelerated administrative management but also ensured document privacy through an automated data security system.Keywords: Digital Archive; Laravel; Document Security; AES-256; Information Systems. AbstrakHambatan dalam efisiensi penemuan kembali dokumen serta tingginya risiko keamanan data pada Unit PMB STIE BPD Jateng menjadi latar belakang dilakukannya penelitian ini. Peneliti mengusulkan sebuah solusi berupa platform kearsipan berbasis web yang dikembangkan menggunakan framework Laravel dan diperkuat dengan skema enkripsi Advanced Encryption Standard (AES-256). Prosedur pengembangan sistem ini mengacu pada metodologi Waterfall yang meliputi fase identifikasi kebutuhan, perancangan struktur, hingga tahap pengkodean. Temuan penelitian menunjukkan bahwa pemanfaatan pola MVC pada Laravel mampu menghasilkan sistem yang lebih terorganisir, sementara integrasi algoritma AES-256 secara efektif memproteksi dokumen dari akses yang tidak memiliki otoritas. Berdasarkan hasil pengujian black-box, seluruh komponen fungsional sistem terbukti beroperasi dengan optimal. Dapat disimpulkan bahwa implementasi teknologi ini tidak hanya mempercepat tata kelola administrasi secara signifikan, tetapi juga menjamin aspek privasi dokumen melalui sistem pengamanan data yang otomatis. 
Prediksi Kadar Air Greenbeans Kopi Pra-Roasting Menggunakan Metode ANFIS Naddiyanto, Muchammad Fadika; Idhom, Mohammad; Maulana, Hendra
Jutisi : Jurnal Ilmiah Teknik Informatika dan Sistem Informasi Vol 15, No 2 (2026): April 2026
Publisher : STMIK Banjarbaru

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35889/jutisi.v15i2.3568

Abstract

Moisture content of green coffee beans is a critical parameter that determines quality stability during storage and the pre-roasting stage; however, conventional measurement methods are destructive and unsuitable for continuous monitoring. This study aims to develop an Internet of Things (IoT)-based moisture content prediction system using the Adaptive Neuro-Fuzzy Inference System (ANFIS). Input variables include temperature, relative humidity (RH), and capacitive sensor ADC signals, while moisture content is used as the target variable. A dataset consisting of 1032 observations was divided into training and testing sets with an 80:20 ratio. The ANFIS model employed Gaussian membership functions and an early stopping mechanism, and its performance was evaluated using MAE, RMSE, MAPE, and the coefficient of determination (R²). Experimental results achieved MAE of 0.2648, RMSE of 0.4187, MAPE of 2.077%, and R² of 0.8109 with an accuracy of 97.923%. The proposed system enables accurate, non-destructive, and real-time moisture content prediction.Keywords: Moisture content; Green beans; Coffee; ANFIS; Prediction.AbstrakKadar air biji kopi hijau merupakan parameter penting yang menentukan stabilitas mutu selama penyimpanan hingga tahap pra-roasting, namun metode pengukuran konvensional bersifat destruktif dan tidak mendukung monitoring berkelanjutan. Penelitian ini bertujuan mengembangkan sistem prediksi kadar air berbasis Internet of Things (IoT) menggunakan metode Adaptive Neuro-Fuzzy Inference System (ANFIS). Variabel input meliputi suhu, kelembaban relatif (RH), dan sinyal ADC sensor, dengan kadar air sebagai variabel target. Dataset sebanyak 1032 data dibagi menjadi data latih dan data uji dengan rasio 80:20. Model ANFIS menggunakan fungsi keanggotaan Gaussian dan mekanisme early stopping, serta dievaluasi menggunakan MAE, RMSE, MAPE, dan koefisien determinasi (R²). Hasil pengujian menunjukkan MAE 0,2648, RMSE 0,4187, MAPE 2,077%, dan R² sebesar 0,8109 dengan akurasi 97,923%. Sistem yang diusulkan mampu melakukan prediksi kadar air secara akurat, non-destruktif, dan real-time. 
Weighted Moving Average Berbasis Variasi Window untuk Optimasi Persediaan dan Reorder Point Fiber Optik Fahturohman, Ridho Fajar; Nugroho, Budi; Puspaningrum, Eva Yulia
Jutisi : Jurnal Ilmiah Teknik Informatika dan Sistem Informasi Vol 15, No 2 (2026): April 2026
Publisher : STMIK Banjarbaru

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35889/jutisi.v15i2.3592

Abstract

Poor inventory management is a primary challenge for telecommunications service providers. This study implements the weighted moving average (WMA) algorithm with window size variations to forecast demand for kabel drop core and Optical Network Terminal (ONT) at Fibertrust Madiun, integrating it with Safety Stock (SS) and Reorder Point (ROP) calculations. The data covers 63 kabel drop core transactions and 112 ONT transactions during January–December 2025, with coefficients of variation of 0.384 and 0.567 respectively. Three window sizes (3, 4, and 5 periods) were tested using MAD, MSE, and MAPE. Window 5 achieved the best accuracy with MAPE of 27.11% for kabel drop core and 47.57% for ONT, both in the sufficient category. A 95% service level provides the optimal balance between holding cost and stockout risk. ROP implementation has the potential to reduce stockout incidents by 71–79%, from 28 to 6–8 incidents per year.Keywords: Weighted moving average; Reorder point; Safety stock; Inventory management; StockoutAbstrakPengelolaan persediaan yang tidak tertata dengan baik menjadi tantangan utama perusahaan penyedia layanan telekomunikasi. Penelitian ini mengimplementasikan algoritma weighted moving average (WMA) dengan variasi ukuran window untuk meramalkan permintaan kabel drop core dan Optical Network Terminal (ONT) di Fibertrust Madiun, serta mengintegrasikannya dengan perhitungan Safety Stock (SS) dan Reorder Point (ROP). Data mencakup 63 transaksi kabel drop core dan 112 transaksi ONT selama Januari-Desember 2025, dengan koefisien variasi masing-masing 0,384 dan 0,567. Tiga variasi ukuran window (3, 4, dan 5 periode) diuji menggunakan metrik MAD, MSE, dan MAPE. Window 5 periode mencatat akurasi terbaik dengan MAPE 27,11% untuk kabel drop core dan 47,57% untuk ONT, keduanya berkategori cukup. Service level 95% memberikan keseimbangan terbaik antara biaya penyimpanan dan risiko kehabisan stok (stockout). Penerapan ROP berpotensi menekan insiden stockout hingga 71-79%, dari 28 insiden menjadi 6-8 insiden per tahun. 
Perancangan Dan Implementasi Sistem IoT Untuk Monitoring Kualitas Air Menggunakan ESP8266 Dan Sensor TDS Meter Darinding, Piki; Setyawan, Gogor Chrismass; Lase, Kristian Juri Dama
Jutisi : Jurnal Ilmiah Teknik Informatika dan Sistem Informasi Vol 15, No 2 (2026): April 2026
Publisher : STMIK Banjarbaru

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35889/jutisi.v15i2.3510

Abstract

Continuous monitoring of water quality plays an important role in supporting effective water resource management. This research develops a water quality monitoring system based on the Internet of Things (IoT) using an ESP8266 microcontroller integrated with a Total Dissolved Solids (TDS) sensor. The study includes hardware design, software development, system integration, and sensor accuracy evaluation by comparing the measurement results with a standard TDS meter. Testing was conducted on three water samples: bottled water, reverse osmosis (RO) gallon water, and tap water. The experimental results show measurement errors of 3.45%, 6.12%, and 6.67%, respectively. The average error obtained is 5.41%, indicating an overall system accuracy of 94.59%. In addition, the system is capable of measuring TDS values in real-time and transmitting the data to the server in a stable manner. These results indicate that the developed system is feasible for IoT-based water quality monitoring applications.Keywords: Internet of Things; Water Quality; ESP8266; TDS Sensor; Real-time Monitoring AbstrakPemantauan kualitas air secara berkelanjutan memiliki peran penting dalam mendukung pengelolaan sumber daya air yang efektif. Penelitian ini mengembangkan sistem monitoring kualitas air berbasis Internet of Things (IoT) dengan memanfaatkan mikrokontroler ESP8266 yang terintegrasi dengan sensor Total Dissolved Solids (TDS). Metode penelitian meliputi perancangan perangkat keras, pengembangan perangkat lunak, integrasi sistem, serta evaluasi akurasi sensor melalui perbandingan hasil pengukuran dengan alat ukur TDS meter. Pengujian dilakukan pada tiga sampel air, yaitu air kemasan, air galon reverse osmosis (RO), dan air keran. Hasil pengujian menunjukkan tingkat kesalahan sebesar 3,45%, 6,12%, dan 6,67%. Rata-rata kesalahan sistem sebesar 5,41% dengan tingkat akurasi mencapai 94,59%. Sistem juga mampu melakukan pengukuran nilai TDS secara real-time dan mengirimkan data ke server secara stabil. Dengan demikian, sistem yang dikembangkan layak digunakan untuk monitoring kualitas air berbasis IoT. 
Penerapan Quantum Machine Learning Untuk Klasifikasi Ulasan Asli Dan Palsu Pada Amazon Telaumbanua, Krisna Putri; Berutu, Sunneng Sandino; Astuti, Aninda
Jutisi : Jurnal Ilmiah Teknik Informatika dan Sistem Informasi Vol 15, No 2 (2026): April 2026
Publisher : STMIK Banjarbaru

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35889/jutisi.v15i2.3498

Abstract

Classifying product reviews on e-commerce platforms, both real and fake, requires a model that can effectively represent text data patterns. This study aims to compare the performance of several Quantum Machine Learning methods, namely QNN, QSVC, and Hybrid Quantum kernel and Classical SVM, in classifying Amazon product reviews. The study uses a quantitative approach with a computational experimental design. Review data is represented using TF-IDF, standardized, and reduced in dimension with Principal Component Analysis (PCA) before being transformed into quantum feature space. Performance evaluation is carried out using accuracy, precision, recall, F1-Score, and MCC metrics. The experimental results show that QNN achieved the best performance with an accuracy value of 85,63%, an F1-Score of 0.9130, and an MCC of 0.5608, while QSVC and the hybrid approach achieved an accuracy of 83.23% with an MCC of 0.4331. These results indicate that QNN has more balanced classification performance.Keywords: Quantum Neural Network; Fake Review Detection; Amazon Reviews; Natural Language Processing; Quantum Machine Learning. AbstrakKlasifikasi ulasan produk pada platform e-commerce, baik ulasan asli maupun palsu, memerlukan model yang mampu merepresentasikan pola data teks secara efektif. Penelitian ini bertujuan untuk membandingkan kinerja beberapa metode Quantum Machine Learning (QML), yaitu QNN, Quantum Support Vector (QSVC), dan Hybrid Quantum kernel and Classical SVM, dalam mengklasifikasikan ulasan produk Amazon. Penelitian menggunakan pendekatan kuantitatif dengan desain eksperimen komputasional. Data ulasan direpresentasikan menggunakan TF-IDF, distandardisasi, dan direduksi dimensinya dengan Principal Component Analysis (PCA) sebelum ditransformasikan ke ruang fitur kuantum. Evaluasi kinerja dilakukan menggunakan metrik accuracy, precision, recall, F1-Score, dan MCC. Hasil eksperimen menunjukkan bahwa QNN memperoleh kinerja terbaik dengan nilai accuracy sebesar 85,63%, F1-Score 0.9130, dan MCC 0.5043, sedangkan QSVC dan pendekatan hybrid mencapai accuracy 83,23% dengan MCC 0,4331. Hasil ini menunjukkan bahwa QNN memiliki performa klasifikasi yang lebih seimbang. 

Page 102 of 103 | Total Record : 1023

 99   100   101   102   103 

Filter by Year

2013 2026


Reset
Filter By Issues
All Issue Vol 15, No 2 (2026): April 2026 Vol 15, No 1 (2026): Februari 2026 Vol 14, No 3: Desember 2025 Vol 14, No 2: Agustus 2025 Vol 14, No 1: April 2025 Vol 13, No 3: Desember 2024 Vol 13, No 2: Agustus 2024 Vol 13, No 1: April 2024 Vol 12, No 3: Desember 2023 Vol 12, No 2: Agustus 2023 Vol 12, No 1: April 2023 Vol 11, No 3: Desember 2022 Vol 11, No 2: Agustus 2022 Vol 11, No 1: April 2022 Vol 10, No 3: Desember 2021 Vol 10, No 2: Agustus 2021 Jutisi Vol. 10, No. 1, April 2021 JUTISI Vol. 9, No. 3, Desember 2020 JUTISI Vol. 9, No. 2, Agustus 2020 JUTISI Vol. 9, No. 1, April 2020 JUTISI Vol. 8, No. 3, Desember 2019 JUTISI Vol. 8, No. 2, Agustus 2019 JUTISI Vol. 8, No. 1, April 2019 Vol 7, No 3: Desember 2018 Vol 7, No 2: Agustus 2018 Vol 7, No 1: April 2018 Vol 6, No 3: Desember 2017 Vol 6, No 2: Agustus 2017 Vol 6, No 1: April 2017 Vol 5, No 3: Desember 2016 Vol 5, No 2: Agustus 2016 Vol 5, No 1: April 2016 Vol 4, No 3: Desember 2015 Vol 4, No 2: Agustus 2015 Vol 4, No 1: April 2015 Vol 3, No 3: Desember 2014 Vol 3, No 2: Agustus 2014 Vol 3, No 1: April 2014 Vol 2, No 1: April 2013 More Issue