cover
Contact Name
Indra
Contact Email
indra@budiluhur.ac.id
Phone
+628568287734
Journal Mail Official
skanika@budiluhur.ac.id
Editorial Address
Jl. Ciledug Raya, Petukangan Utara, Jakarta Selatan, Jakarta Selatan, Provinsi DKI Jakarta, 12260
Location
Kota adm. jakarta selatan,
Dki jakarta
INDONESIA
SKANIKA: Sistem Komputer dan Teknik Informatika
ISSN : -     EISSN : 27214788     DOI : 10.36080
SKANIKA: Sistem Komputer dan Teknik Informatika adalah media publikasi online hasil penelitian yang diterbitkan oleh Program Studi Sistem komputer dan Teknik Informatika, Fakultas Teknologi Informasi, Universitas Budi Luhur. Scope atau Topik Jurnal: Kriptografi, Steganografi, Sistem Pakar / Artificial Intelligence , Sistem Penunjang Keputusan, Bioinformatika, Kecerdasan Komputasional, Semantics Web dan Ontologies, Data Mining,Text Mining,Natural Language Processing, Pengelolaan Citra Digital, Otomasi Berbasis Sensor, Wireless Sensor Network, Network Management dan Maintenance, Sistem Operasi, Sosial Network Analysis, Security, Augmented Reality, Game Development, Virtual Reality, Webservice / API, Internet of Things (IoT)
Articles 340 Documents
Pengembangan Sistem Rekomendasi Berpakaian Menggunakan Local Binary Pattern dan K-Nearest Neighbor di Program Studi Teknik Informatika Universitas Pelita Bangsa M. Najamuddin Dwi Miharja; Helmi Ahmad Fauzi Candra; Nanang Tedi
SKANIKA: Sistem Komputer dan Teknik Informatika Vol 8 No 2 (2025): Jurnal SKANIKA Juli 2025
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/skanika.v8i2.3584

Abstract

The student code of ethics governs behavior, speech, actions, appearance, and dress during their academic journey. At Pelita Bangsa University, many students tend to follow evolving fashion trends, which often conflict with the faculty's dress code that emphasizes wearing formal, collared clothing. This research addresses the issue by developing an image-based detection system to identify whether students wear formal or informal attire. The study utilizes the Local Binary Pattern (LBP) method for feature extraction and the K-Nearest Neighbor (K-NN) method for classification. A total of 130 images were tested, consisting of 70 t-shirts and 60 shirts. The best accuracy was achieved using parameters R=1 and P=8 for LBP and K=1 with Euclidean distance for K-NN, resulting in an average accuracy of 95.16%. The developed system is capable of accurately classifying images of t-shirts and shirts, demonstrating high precision and efficiency in image-based classification. These findings indicate that the application of the Local Binary Pattern (LBP) and K-Nearest Neighbor (K-NN) methods is an effective combination for detecting compliance with student dress code regulations
Implementasi Arsitektur Recurrent Neural Network Pada Analisis Sentimen Clash of Champions Arif Hidayat; Anindita Septiarini; Medi Taruk
SKANIKA: Sistem Komputer dan Teknik Informatika Vol 8 No 2 (2025): Jurnal SKANIKA Juli 2025
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/skanika.v8i2.3586

Abstract

Clash of Champions is an educational program by Ruangguru on YouTube that has received mixed responses. This study aims to perform sentiment analysis using three Recurrent Neural Network (RNN) architectures: Vanilla Recurrent Neural Network (Vanilla RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU). The data consists of 2,100 training samples, 300 validation samples, and 600 testing samples collected from YouTube and enriched with data augmentation using GPT-4 technology. Additionally, 35 comments from a survey conducted via Google Form are used for generalization testing. Comments are classified into three sentiments: Pro, Neutral, and Contra. The analysis involves preprocessing, model training, and evaluation using standard metrics. GRU demonstrated the best performance with an accuracy of 99.2% and the highest F1 score. LSTM achieved an accuracy of 99.0% and a recall of 100% for the Pro class, while Vanilla RNN was less stable. On real-world data, GRU correctly predicted 16 comments, outperforming LSTM with 14 correct predictions and RNN with 13 correct predictions. GRU excels in accuracy, stability, and adaptability to the data.
Penerapan Metode Fuzzy Tsukamoto Untuk Penentuan Negara Penyuplai Beras Pada Perusahaan Logistik Eneng Siti Nurjanah; Irmayansyah Irmayansyah; Leny Tritanto Ningrum
SKANIKA: Sistem Komputer dan Teknik Informatika Vol 8 No 2 (2025): Jurnal SKANIKA Juli 2025
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/skanika.v8i2.3588

Abstract

Inaccuracy in determining rice supplier countries is currently a complex issue. This situation is influenced by various factors that are often difficult to predict and have the potential to suddenly change rice supply conditions. Extreme climate change and unexpected fluctuations in domestic demand can also influence decisions regarding rice supplier countries. Under these conditions, logistics companies require a more adaptive and responsive supply strategy to changes in rice availability. Selecting the right supplier country plays a crucial role in the operational efficiency and business sustainability of logistics companies. This study aims to apply the Fuzzy Tsukamoto method as a supporting tool in the process of determining rice supplier countries for logistics companies. The Tsukamoto method is used to calculate rice supplier categories by utilizing three input variables: import quantity, demand quantity, and rice quality. Meanwhile, the output variables are classified into three categories: low, medium, and high. The accuracy of the rice supplier country grouping results using the Tsukamoto method has been tested and obtained an accuracy value of 73.33%. The developed application prototype has also undergone user feasibility testing using the Post-Study System Usability Questionnaire (PSSUQ) instrument and resulted in a feasibility level of 91.1%.
Implementasi Jaringan Syaraf Tiruan Backpropagation untuk Peramalan Penjualan pada PT. Central Pacific Development Indra Hertanto; Riskiana Wulan; Lutfi Rizaldi Mahida; Dzaky Rakha Meilano; Prayoga Ajitya Setiawan; Indra Indra
SKANIKA: Sistem Komputer dan Teknik Informatika Vol 8 No 2 (2025): Jurnal SKANIKA Juli 2025
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/skanika.v8i2.3589

Abstract

Central Pacific Development possesses an abundance of sales transaction data, yet currently lacks a system to optimally leverage this data for strategic planning. This research aims to implement an Artificial Neural Network (ANN) using the Backpropagation method to predict product sales, based on historical data from January 2022 to November 2024. The system involves stages such as data normalization, splitting the dataset into training and testing sets, and evaluating model performance using Mean Squared Error (MSE) and Mean Absolute Percentage Error (MAPE) metrics. A Multi-Layer Perceptron (MLP) model with a 12-15-1 configuration yielded the best results, achieving a training MSE of 0.000999, a testing MSE of 0.062680, a MAPE of 22.24%, and an accuracy of 77.75%. The developed system can assist the company in designing data-driven production and marketing strategies, while also opening opportunities for further development through the integration of big data technologies or hybrid methods to improve prediction accuracy.
IMPLEMENTASI MEL-FREQUENCY CEPSTRAL COEFFICIENTS DAN CONVOLUTIONAL NEURAL NETWORK UNTUK PENGENALAN HURUF HIRAGANA Muhammad Yusuf Ibrahim Ramadhani; Saiful Nur Budiman; Udkhiati Mawaddah
SKANIKA: Sistem Komputer dan Teknik Informatika Vol 9 No 1 (2026): Jurnal SKANIKA Januari 2026
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/skanika.v9i1.3600

Abstract

Japanese language learning has gained increasing interest in Indonesia; however, learners often experience difficulties in mastering Hiragana characters due to their large number and phonetic similarities. Speech recognition technology can be utilized as a supportive learning medium, particularly for improving pronunciation and enhancing learners’ understanding of Hiragana characters. This study aims to develop a Hiragana speech recognition system based on Mel-Frequency Cepstral Coefficients (MFCC) for feature extraction and Convolutional Neural Networks (CNN) for classification. The dataset consists of 46 Hiragana characters, with each character recorded 20 times by four speakers, resulting in a total of 3,680 audio samples. The research stages include audio signal preprocessing, MFCC feature extraction, data augmentation, CNN model training, and performance evaluation using classification metrics. Experimental results indicate that the proposed model achieves an accuracy of 95% on the test data, with most Hiragana characters being correctly recognized. Misclassifications mainly occur among characters with similar phonetic characteristics. These results demonstrate that the MFCC-based CNN approach is effective for Hiragana speech recognition and has potential to be applied as an interactive digital learning medium for Japanese language education.
KLASIFIKASI KOMENTAR NETIZEN X TENTANG PEMECATAN SHIN TAE YONG DARI PSSI DENGAN ALGORITMA NAIVE BAYES Hidayat, Rizky; Windarto, Windarto
SKANIKA: Sistem Komputer dan Teknik Informatika Vol 9 No 1 (2026): Jurnal SKANIKA Januari 2026
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/skanika.v9i1.3605

Abstract

The dismissal of the Indonesian National Football Team head coach, Shin Tae-yong, generated diverse public reactions on the social media platform X. The large volume and variability of netizen comments require a systematic analysis to objectively understand public opinion. This study aims to analyze the sentiment of netizen comments regarding the dismissal of Shin Tae-yong using a text mining approach and the Multinomial Naive Bayes algorithm. The data were collected from social media X through a crawling process and subsequently processed through preprocessing stages and TF-IDF weighting. The classification results demonstrate that the proposed model achieved good performance, with an accuracy of 91.3%, precision of 94.94%, recall of 79.43%, and an F1-score of 85.4%. The prediction results were dominated by negative sentiments (275 instances), followed by positive (40 instances) and neutral sentiments (30 instances). These findings indicate that public opinion tends to be predominantly negative toward the decision, while the classification model effectively categorizes sentiments. This study is expected to serve as a reference for understanding public opinion in national sports issues and to support data-driven decision-making.
ANALISIS SENTIMEN ULASAN PRODUK SPAREPART MOTOR DI E-COMMERCE MENGGUNAKAN METODE SUPPORT VECTOR MACHINE (SVM) Reza Ega Resnanda; Dwi Cahyono; Anik Vega Vitianingsih
SKANIKA: Sistem Komputer dan Teknik Informatika Vol 9 No 1 (2026): Jurnal SKANIKA Januari 2026
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/skanika.v9i1.3608

Abstract

This study was motivated by the increasing use of e-commerce in Indonesia, which highlights the importance of analyzing customer reviews as a basis for evaluating product and service quality. This study aims to analyze the sentiment of reviews of Honda motorcycle spare parts at the Ducks Garage store on the Tokopedia platform using the Support Vector Machine (SVM) algorithm. The dataset used consists of 2.537 reviews obtained through web scraping techniques and processed through text preprocessing stages, including data cleaning, normalization, tokenization, stopword removal, and stemming. Sentiment labelling was carried out into three classes, namely positive, negative, and neutral, with lexicon-based and feature weighting using the Term Frequency–Inverse Document Frequency (TF-IDF) method. Data distribution imbalance was handled using the Synthetic Minority Over-Sampling Technique (SMOTE) method. The SVM model was tested using three types of kernels, namely Linear, Polynomial, and Radial Basis Function (RBF). The test results showed that the RBF kernel produced the best performance with an accuracy of 92.79%, followed by the Linear kernel at 89.89% and the Polynomial kernel at 72.57%. The conclusion of this study shows that the application of SVM with SMOTE data balancing is effective in classifying the sentiment of e-commerce product reviews and can be used to support data-driven business decisions based on customer data.
RANCANG BANGUN SISTEM PERAMALAN PENJUALAN ES KRIM DENGAN METODE DOUBLE EXPONENTIAL SMOOTHING (STUDI KASUS :CV ALDY ABADI JAYA, MUARA WAHAU) Viramadita Arthamia Putri; Sentot Achmadi; Ali Mahmudi
SKANIKA: Sistem Komputer dan Teknik Informatika Vol 9 No 1 (2026): Jurnal SKANIKA Januari 2026
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/skanika.v9i1.3618

Abstract

The ice cream sales agent CV Aldy Abadi Jaya in Muara Wahau faces challenges in forecasting daily fluctuating ice cream demand. Inaccurate stock estimation often leads to excess or shortage of inventory, requiring an effective strategy to balance supply and demand. This study aims to design and develop a web-based sales forecasting system using the Double Exponential Smoothing (DES) method. This method was chosen because it can provide accurate predictions for data with a trend pattern. The research utilizes historical sales data of ice cream from CV Aldy Abadi Jaya. The system was developed using the Laravel framework and MySQL database. It automatically performs sales forecasting calculations and displays the results through an interactive dashboard that helps users interpret prediction outcomes easily. The results show an accuracy level and Mean Absolute Percentage Error (MAPE) value of 13,83%, indicating that the Double Exponential Smoothing method can generate reliable forecasts for future sales. The developed system is expected to help the company plan stock more effectively, reduce the risk of losses, and improve efficiency in the ice cream distribution process..
ANALISIS SENTIMEN TERHADAP KOMENTAR PADA VIDEO VIRAL (FYP) TIKTOK MENGGUNAKAN METODE NAÏVE BAYES: Indonesia Pratama, Putra Weka; Endang Wahyu Pamungkas
SKANIKA: Sistem Komputer dan Teknik Informatika Vol 9 No 1 (2026): Jurnal SKANIKA Januari 2026
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/skanika.v9i1.3628

Abstract

Sentiment analysis, or opinion mining, is a method used to identify public opinions expressed in textual form on social media platforms. This approach is useful for understanding public responses to a particular phenomenon without the need for conventional survey methods. This study focuses on sentiment analysis of user comments on viral TikTok videos categorized under the For You Page (FYP). The dataset was obtained through a comment crawling process, resulting in 12,494 comments, which were then processed through preprocessing stages including case folding, text normalization, stopword removal, and stemming. Sentiment classification was performed using the Naïve Bayes method with Term Frequency–Inverse Document Frequency (TF-IDF) weighting and two sentiment classes, namely positive and negative. The data were split using an 80% training and 20% testing scheme. Experimental results show that the proposed method achieved a best accuracy of 90%, demonstrating that the combination of comprehensive preprocessing and TF-IDF weighting effectively improves the performance of sentiment classification on comments from TikTok FYP videos.
SIG DENGAN K-MEANS++ UNTUK KLASTERISASI PENGEMBANGAN UMKM KAIN TENUN (STUDI KASUS: KABUPATEN NAGEKEO) Wulang, Maria Yasinta; Wibowo, Suryo Adi; Susanto, Eko Heri
SKANIKA: Sistem Komputer dan Teknik Informatika Vol 9 No 1 (2026): Jurnal SKANIKA Januari 2026
Publisher : Universitas Budi Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36080/skanika.v9i1.3630

Abstract

The woven cloth Small and Medium Enterprises (SMEs) in Nagekeo Regency possess significant economic and cultural potential; however, the current coaching process is executed uniformly without data-driven analysis, resulting in inefficient allocation of aid. This study aims to map the distribution of woven cloth SMEs, develop a web-based Geographic Information System (GIS) application, and implement the K-Means++ method to cluster the SMEs based on their productivity levels. The system was designed using Laravel and Leaflet.js, incorporating features for data management, interactive maps, and visualization of productivity clusters, which include Medium Productivity (PM), Low Productivity (PR), and Dense/Massive Productivity (PP). The research findings indicate that the system's clustering process achieved 100% accuracy compared to manual calculation using Excel, with a 0% error rate. A lift ratio of 7.69 (>1) signifies a strong relationship between variables and validates the clustering results. The algorithm's computation time was recorded at 0.464 seconds. Black-box and browser compatibility tests confirmed that all features functioned as intended across Chrome, Edge, and Firefox. Furthermore, user testing involving 10 respondents yielded a positive assessment, with percentages of 43% Strongly Agree, 41% Agree, 14.5% Neutral, and 1.5% Disagree. This system is capable of supporting more effective and objective spatial data-driven decision-making