cover
Contact Name
Hidayat
Contact Email
hidayat@email.unikom.ac.id
Phone
-
Journal Mail Official
tk@email.unikom.ac.id
Editorial Address
-
Location
Kota bandung,
Jawa barat
INDONESIA
KOMPUTIKA - Jurnal Sistem Komputer
ISSN : 22529039     EISSN : 26553198     DOI : -
Jurnal Ilmiah KOMPUTIKA adalah wadah informasi berupa hasil penelitian, studi kepustakaan, gagasan, aplikasi teori dan kajian analisis kritis di bidang kelimuan bidang Sistem Komputer.
Arjuna Subject : -
Articles 218 Documents
Perbandingan kualitas hubungan Leader - member exchange terhadap Followers Outcome pada perkembangan digital startup Tuzzahra, Zabrina; ER, Mahendrawati
Komputika : Jurnal Sistem Komputer Vol. 14 No. 1 (2025): Komputika: Jurnal Sistem Komputer
Publisher : Computer Engineering Departement, Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/komputika.v14i1.15662

Abstract

The quality of relationships between employees in a growing company greatly influences increasing followers outcomes, this is also inseparable from the role of a leader or team leader. The leader-member exchange (LMX) theory is a leadership theory that considers the dyadic relationship between leaders and team members, where each relationship quality certainly has different follower outcomes. The purpose of this study was to analyze the differences in the quality of relationships that occur between team leaders and team members during organizational change in each phase. The data used in the study were obtained from a digital startup company in East Java, in 3 different divisions in each phase and processed using the case study research method. The results of the study prove that there are differences in the quality of relationships in the first phase, the quality of the relationship is classified as negative, while in the second phase it is positive. And an anomalous phenomenon was found in the quality of the relationships analyzed.
Prediksi Jumlah Migrasi Penduduk dengan Menerapkan Jaringan Syaraf Menggunakan Metode Backpropagation suhendro, dedi; Adriatasya, Sabila Putri
Komputika : Jurnal Sistem Komputer Vol. 14 No. 1 (2025): Komputika: Jurnal Sistem Komputer
Publisher : Computer Engineering Departement, Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/komputika.v14i1.15768

Abstract

Indonesia is a country that has the fourth highest population density in the world, the distribution of Indonesia's population is not evenly distributed for each region which finally the government took a policy to carry out a migration program from one region to another, thus affecting the economic development of Indonesia's population. When choosing a place to live, there are various factors that must be considered. If someone is not suitable to settle in an area or there are other driving factors, they can move or commonly known as migration. This research aims to predict the number of migrations by analyzing the results using Artificial Neural Networks with the Backpropagation Method. The Backpropagation method can be utilized by providing a relationship between population data so that it is expected in this study to achieve the right prediction. The best modeling is obtained in the 6-16-2 architecture using an input layer of 6 neurons, a hidden layer of 16 neurons and an output layer of 1 neuron. Architecture 6-16-1 produces an epoch of 8939 iterations, time 01:42, training MSE 0.00190005 and testing MSE 0.05752814 with an accuracy rate of 82%. Keywords – Prediction; Migration; Population; Artificial Neural Network; Backpropagation Method.
Perancangan, Fabrikasi, dan Karakterisasi Transistor Bipolar: Perancangan, Fabrikasi, dan Karakterisasi Transistor Bipolar Subandi, Ayub; Idris, Irman
Komputika : Jurnal Sistem Komputer Vol. 14 No. 1 (2025): Komputika: Jurnal Sistem Komputer
Publisher : Computer Engineering Departement, Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/komputika.v14i1.15859

Abstract

Bipolar transistors still play an important and irreplaceable role in the application of electronic systems, even though almost all logic circuits, including microprocessors and memory, consist of MOS (Metal Oxide Semiconductor) transistors. They are widely used today due to their advantage higher switching speed compared to MOS technology and their ability to conduct high currents. In this research, the design, fabrication, and characterization of bipolar transistor were carried out. The design was done using L-Edit software and the converted to glass using a photoreduction camera. The base material for the transistors was n-type epitaxial wafer with a layer depth of 3 μm. The fabrication process involved wafer cleaning, oxidation, diffusion, photolithography, etching, metallization, and metal strengthening. Due to the use of a thin epitaxial wafer, thermal diffusion followed by a drive-in process was avoided to prevent penetration through the epitaxial layer, resulting in a low current gain (β) of only 1.5 times. On the other hand, using a thicker epitaxial layer would create difficulties during the isolation process. The success of each fabrication step was determined by measurement on test patterns, including resistor resistance measurement, diode characterization, and determination of diffusion depth using dummy samples. The measurement results showed an emitter junction depth of 0.3 μm, and a base layer width of 1.1 μm. These results indicate that the transistor was successfully fabrication, although with a low current gain. Keywords – MOS; Epitaxial wafer; Diffusion; Current amplifier; Dummy. 
Sistem Pengawasan Keamanan Otomatis dengan Sound Sensor untuk Mencegah Bullying: Sistem Pengawasan Keamanan Otomatis dengan Sound Sensor untuk Mencegah Bullying Shofiyullah, R. Muhammad Azmi Herdi
Komputika : Jurnal Sistem Komputer Vol. 14 No. 2 (2025): Komputika: Jurnal Sistem Komputer
Publisher : Computer Engineering Departement, Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/komputika.v14i2.13932

Abstract

Perundungan merupakan masalah serius di berbagai lingkungan, termasuk di tempat kerja dan sekolah, yang membutuhkan perhatian dan tindakan pencegahan. Di Indonesia, tingkat perundungan di kalangan pelajar mencapai 41%, dengan dampak negatif seperti rendah diri, depresi, dan kesulitan belajar. Untuk mengatasi masalah ini, diperlukan solusi yang meningkatkan kualitas pengawasan keamanan guna mengurangi tingkat perundungan. Salah satu langkah yang dapat diambil adalah menggunakan perangkat yang mampu melakukan pengawasan secara otomatis. Penggunaan Internet of Things (IoT) dengan sensor suara berbasis metode machine learning, seperti Support Vector Machine (SVM), dapat meningkatkan efisiensi pengawasan dan mendeteksi perundungan secara real-time dengan memanfaatkan fitur suara teriakan. Dengan menerapkan metode SVM dan didukung dengan IoT, pengawasan keamanan dapat dilakukan secara otomatis. Dalam penelitian ini, hasil model klasifikasi teriakan dengan menggunakan metode SVM memiliki akurasi sebesar 98% pada saat training, 76% pada saat pelaksanaan testing dengan menggunakan dataset, dan 55% pada saat percobaan langsung dengan menggunakan sensor KY-037. Hasil tersebut menunjukkan bahwa sensor suara KY-037 dapat digunakan untuk melakukan pengklasifikasian suara dengan bantuan machine learning. Kata Kunci – IoT; SVM; Machine Learning; Keamanan; Perundungan. 
Optimasi Prediksi Kelayakan Pinjaman dengan Teknik Resampling dan Algoritma Boosting Putra, Muhammad Ricky Perdana; Juwariyah, Siti; Ridwan, Muhammad; Marco, Robert
Komputika : Jurnal Sistem Komputer Vol. 14 No. 2 (2025): Komputika: Jurnal Sistem Komputer
Publisher : Computer Engineering Departement, Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/komputika.v14i2.15485

Abstract

Loan eligibility assessment is a crucial element in financial risk mitigation, aiming to minimize potential losses due to bad debts and ensure proper resource distribution. Traditional rule-based approaches have limitations in scalability, risk of subjective bias, and complex data management. The application of Machine Learning (ML) presents a solution with the ability to analyze complex patterns in historical data, although significant challenges such as class imbalance where the number of defaulted borrowers is much smaller than that of current borrowers and missing values ​​in the dataset remain major obstacles. This study evaluates the SMOTE and SMOTE-ENN resampling methods, to address class imbalance, as well as the mean imputation technique to handle missing values. By evaluating boosting algorithms, including Gradient Boosting, XGBoost, LightGBM, AdaBoost, and CatBoost, the results show that the combination of the CatBoost algorithm with the SMOTE-ENN sampling technique provides the highest prediction accuracy of 91.67%. This finding confirms the significant potential of ML in improving the accuracy, efficiency, and fairness of predictions, while making important contributions to the development of data-driven decision-making systems in the financial sector.
Peningkatan Kualitas Gambar Wajar Pada Sistem Deteksi Wajah Menggunakan GFPGAN Sugeng, Sugeng; Mansyur, Moh Ripan
Komputika : Jurnal Sistem Komputer Vol. 14 No. 2 (2025): Komputika: Jurnal Sistem Komputer
Publisher : Computer Engineering Departement, Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/z6n9am93

Abstract

This study investigates the capability of the Generative Adversarial Network (GAN) algorithm, specifically GFPGAN (Generative Face Prior-Generative Adversarial Network), in enhancing the quality of facial images to support more accurate face recognition. GFPGAN is known for its effectiveness in restoring degraded facial images, such as those affected by blurriness, noise, and low resolution—common issues in CCTV (Closed-Circuit Television) footage or other low-quality image sources. By leveraging the GAN architecture, which consists of a generator and a discriminator, GFPGAN is able to produce high-detail facial images while preserving the original facial identity. In this research, GFPGAN is utilized to restore degraded facial images prior to the face recognition process using the DLIB library. Various types of image degradation are tested, including blur, noise, grayscale conversion, and JPEG (Joint Photographic Experts Group) compression. The evaluation involves comparing the face recognition success rate using DLIB before and after the restoration process with GFPGAN. The results demonstrate that previously unrecognizable images become identifiable after being processed with GFPGAN, thereby confirming that image restoration can significantly improve face recognition accuracy.
Segmentasi Pola Pembatalan Pemesanan Layanan di Salon Nail Art XYZ Menggunakan K-Means Clustering dan Evaluasi dengan Davies-Bouldin Index Banusu, Junita Gregoria; Banusu, Junita; Crisintha, Debora; Baso, Budiman; Herlina Ullu, Hevi
Komputika : Jurnal Sistem Komputer Vol. 14 No. 2 (2025): Komputika: Jurnal Sistem Komputer
Publisher : Computer Engineering Departement, Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/komputika.v14i2.15937

Abstract

The beauty industry, particularly nail art, is experiencing rapid growth as consumer demand for nail care services increases. However, the high booking cancellation rate is a challenge that can reduce operational efficiency and customer loyalty. This research focuses on segmenting XYZ Nail Art Salon customers based on cancellation patterns by applying the K-Means algorithm and evaluating using the Davies-Bouldin Index (DBI). In contrast to previous research which is generally conducted in e-commerce, hotels, and retail, this research focuses on the beauty service industry which is still rarely explored. The data used comes from the Kaggle platform with 400 order data. The analysis process includes pre-processing, key feature selection (Cancel Description and Days), and K-Means implementation. The clustering results showed two main segments: 288 customers with planned cancellation patterns and 112 customers with spontaneous cancellation patterns. Evaluation using DBI yielded a value of 0.399 indicating good clustering quality. This segment distinction has practical implications, such as automatic reminder strategies for customers with planned cancellations, as well as providing schedule flexibility or special promotions for customers with spontaneous cancellations. This research contributes to providing data-driven insights for salon management to devise more targeted marketing strategies, reducing the level of customer churn.
Optimization of Production Operator Performance Assessment with Grey Geometric Mean Weighting and Combinative Distance-based Assessment Wang, Junhai; Setiawansyah, Setiawansyah; Ulum, Faruk; Yudhistira, Aditia; Wahyudi, Agung Deni
Komputika : Jurnal Sistem Komputer Vol. 14 No. 2 (2025): Komputika: Jurnal Sistem Komputer
Publisher : Computer Engineering Departement, Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/komputika.v14i2.15977

Abstract

The performance of production operators plays a crucial role in determining the level of efficiency and effectiveness of the manufacturing process in a company that has a long-term impact on the company's competitiveness. Production operator performance appraisals often face a number of problems that can reduce the accuracy and fairness of evaluations. One of the main problems is the subjectivity of assessment, where evaluation is based more on the personal perception of the supervisor or assessor without a consistently measurable standard. The purpose of this study is to apply a more objective, structured, and accurate production operator performance evaluation model by integrating the grey geometric mean weighting (G2M Weighting) method as an uncertainty-based criterion weighting approach and combinative distance-based assessment (CODAS) as an alternative ranking method. The results of the production operator's performance ranking are that CR Operator ranks first with the highest performance score of 0.7737, GM Operator is ranked second with a score of 0.6187, followed by AN Operator in third place with a score of 0.5895. This research makes a significant contribution to the development of a performance evaluation system in the manufacturing industry environment by integrating the G2M Weighting and CODAS methods as an objective and systematic approach.
Analisis K-Means Cluster Kabupaten/Kota di Provinsi Kalimantan Selatan berdasarkan Indikator Indeks Pembangunan Manusia Muhammad Naufal Nor Akmal; Akhmad Yusuf; Ahmad Mishbahul Munier; Muhammad Tezhar Rayhan Noor
Komputika : Jurnal Sistem Komputer Vol. 14 No. 2 (2025): Komputika: Jurnal Sistem Komputer
Publisher : Computer Engineering Departement, Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/komputika.v14i2.16301

Abstract

Indeks Pembangunan Manusia (IPM) merupakan indikator komprehensif yang digunakan untuk mengukur kualitas hidup penduduk melalui tiga dimensi utama: kesehatan, pendidikan, dan ekonomi. Di Provinsi Kalimantan Selatan, masih terjadi kesenjangan capaian IPM antar kabupaten/kota, yang mencerminkan adanya ketimpangan dalam pembangunan manusia. Untuk itu, diperlukan pendekatan analitis guna mengelompokkan wilayah-wilayah berdasarkan IPM-nya. Penelitian ini bertujuan untuk mengelompokkan 13 kabupaten/kota di Provinsi Kalimantan Selatan berdasarkan indikator IPM menggunakan metode K-Means Clustering. Data yang digunakan mencakup empat indikator: Umur Harapan Hidup (UHH), Harapan Lama Sekolah (HLS), Rata-rata Lama Sekolah (RLS), dan Pengeluaran per Kapita. Sebelum clusterisasi, data dinormalisasi menggunakan metode Min-Max Scaling untuk menyamakan skala variabel. Selanjutnya, K-Means diterapkan untuk membentuk cluster berdasarkan kesamaan karakteristik. Pemilihan jumlah cluster optimal dilakukan menggunakan metode Elbow. Hasil analisis menunjukkan bahwa tiga cluster terbentuk secara optimal: cluster dengan IPM sangat tinggi, tinggi, dan sedang. Cluster IPM sangat tinggi terdiri dari Kota Banjarbaru dan Kota Banjarmasin; cluster tinggi mencakup mayoritas kabupaten; sementara cluster sedang hanya mencakup dua kabupaten dengan capaian IPM paling rendah. Temuan ini memberikan gambaran mengenai ketimpangan pembangunan manusia di Kalimantan Selatan. Kontribusi utama penelitian ini adalah menyediakan dasar analitik berbasis data untuk mendukung perumusan kebijakan pembangunan yang lebih efektif, khususnya dalam mengurangi kesenjangan IPM antar wilayah.
Implementasi Metode Random Forest dalam Analisis Prediksi Dogecoin Siregar, Bakti; Jimy, Valensius
Komputika : Jurnal Sistem Komputer Vol. 14 No. 2 (2025): Komputika: Jurnal Sistem Komputer
Publisher : Computer Engineering Departement, Universitas Komputer Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34010/komputika.v14i2.16486

Abstract

Perkembangan teknologi digital telah mendorong munculnya mata uang kripto seperti Dogecoin yang memiliki volatilitas harga tinggi, sehingga menarik perhatian investor namun juga menimbulkan risiko prediksi harga yang kompleks. Penelitian ini menawarkan solusi prediksi harga Dogecoin dengan menerapkan metode Random Forest, salah satu algoritma machine learning berbasis ensemble learning yang unggul dalam menangani data berpola nonlinier. Data yang digunakan adalah data historis harian harga Dogecoin periode 1 Januari–31 Desember 2024 yang diperoleh dari investing.com. Tahapan penelitian meliputi pengunduhan dan pra-pemrosesan data untuk mengatasi missing values, pembagian dataset menjadi data latih (90%) dan data uji (10%), pembangunan model Random Forest menggunakan perangkat lunak R, serta evaluasi kinerja model menggunakan Mean Absolute Percentage Error (MAPE). Hasil penelitian menunjukkan bahwa model mampu memprediksi harga Dogecoin dengan nilai MAPE 2,59% dan R-squared 99,37%, yang menandakan tingkat akurasi tinggi. Kontribusi penelitian ini adalah memberikan model prediksi harga kripto yang andal dan terukur, yang dapat dimanfaatkan investor maupun peneliti sebagai acuan pengambilan keputusan, sekaligus memperluas literatur metode Random Forest pada analisis harga aset digital.