cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Mesran
Contact Email
mesran.skom.mkom@gmail.com
Phone
+6282161108110
Journal Mail Official
ijics.stmikbudidarma@gmail.com
Editorial Address
Jalan Sisingamangaraja No. 338, Simpang Limun, Medan, Sumatera Utara
Location
Kota medan,
Sumatera utara
INDONESIA
The IJICS (International Journal of Informatics and Computer Science)
Published by Sekolah Tinggi Manajemen Informatika dan Komputer Budi Darma
ISSN : 25488449     EISSN : 25488384     DOI : https://doi.org/10.30865/ijics
Core Subject : Science, Engineering,
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering
The The IJICS (International Journal of Informatics and Computer Science) covers the whole spectrum of intelligent informatics, which includes, but is not limited to : • Artificial Immune Systems, Ant Colonies, and Swarm Intelligence • Autonomous Agents and Multi-Agent Systems • Bayesian Networks and Probabilistic Reasoning • Biologically Inspired Intelligence • Brain-Computer Interfacing • Business Intelligence • Chaos theory and intelligent control systems • Clustering and Data Analysis • Complex Systems and Applications • Computational Intelligence and Soft Computing • Cognitive systems • Distributed Intelligent Systems • Database Management and Information Retrieval • Evolutionary computation and DNA/cellular/molecular computing • Expert Systems • Fault detection, fault analysis and diagnostics • Fusion of Neural Networks and Fuzzy Systems • Green and Renewable Energy Systems • Human Interface, Human-Computer Interaction, Human Information Processing • Hybrid and Distributed Algorithms • High Performance Computing • Information storage, security, integrity, privacy and trust • Image and Speech Signal Processing • Knowledge Based Systems, Knowledge Networks • Knowledge discovery and ontology engineering • Machine Learning, Reinforcement Learning • Memetic Computing • Multimedia and Applications • Networked Control Systems • Neural Networks and Applications • Natural Language Processing • Optimization and Decision Making • Pattern Classification, Recognition, speech recognition and synthesis • Robotic Intelligence • Rough sets and granular computing • Robustness Analysis • Self-Organizing Systems • Social Intelligence • Soft computing in P2P, Grid, Cloud and Internet Computing Technologies • Stochastic systems • Support Vector Machines • Ubiquitous, grid and high performance computing • Virtual Reality in Engineering Applications • Web and mobile Intelligence, and Big Data
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 153 Documents
 12   13   14   15   16 
Comparative Study of Naive Bayes and SVM for E-Commerce Sentiment Classification on Shopee Yoga Fradana; Cahyo Adi Nugraha; Frans Nicko Apriansyah; Tri Mutiara Illahi; Ken Ditha Tania; Allsela Meiriza
The IJICS (International Journal of Informatics and Computer Science) Vol. 10 No. 2 (2026): July
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/ijics.v10i2.9740

Abstract

This study compares the performance of Naive Bayes and Support Vector Machine (SVM) for sentiment classification of men’s shirt product reviews on Shopee. A dataset of 500 reviews was collected via web scraping and processed through case folding, tokenizing, stopword removal, and stemming, followed by TF-IDF feature extraction. The data was split at an 80:20 ratio and evaluated using accuracy, precision, recall, and F1-score. The main contribution of this study is demonstrating that despite both algorithms achieving equal overall accuracy of 93%, SVM outperforms Naive Bayes in detecting negative sentiment on a class-imbalanced dataset, with SVM attaining a negative class recall of 0.87 and F1-score of 0.88 compared to 0.80 and 0.87 for Naive Bayes. These findings provide practical guidance for selecting an appropriate classifier in imbalanced e-commerce review classification tasks.
Classification of Coming‑of‑Age Song Lyrics Using Convolutional Neural Network (CNN) Architecture Arjon Samuel Sitio; Fricles Ariwisanto Sianturi; Anita Sindar
The IJICS (International Journal of Informatics and Computer Science) Vol. 10 No. 2 (2026): July
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/ijics.v10i2.9775

Abstract

The coming-of-age theme in song lyrics is rich in emotional expressions, identity transitions, and nostalgia. This study implements a 1-Dimensional (1D) Convolutional Neural Network (CNN) architecture to automatically classify coming-of-age-themed song lyrics. Raw lyrics data were collected through a custom scraping function get_song_lyrics, which then went through a structured text preprocessing stage to remove stopwords and non-semantic components. The cleaned words were represented into a vector space using a pre-trained 50-dimensional GloVe embedding (GloVe 50d) of size (max_length, 50) to capture semantic relationships between words. A 1D CNN model was applied to extract local features in the form of phrase combinations (n-grams) through filter shifting, followed by a Global Max Pooling layer to filter out the most dominant emotional information before the final classification process. As a comparison method and additional analysis, a rule-based approach using TextBlob was applied to extract polarity and subjectivity scores, while the Word Cloud technique was used to visualize the dominance of transitional lexical terms such as grow, leave, and remember. The gap between training and validation performance suggests that the model learned dataset-specific patterns rather than generalized semantic representations. Similar overfitting behavior has been reported in CNN-based lyric and sentiment classification studies when training data are limited or insufficiently diverse. The results showed that the integration of GloVe 50d semantic representation and local feature extraction by 1D CNN was able to produce high and stable accuracy in recognizing the unique characteristics of song lyrics with a maturity theme.
Implementation of K-Means Clustering for Student Achievement Classification Using Academic Performance Indicators Dedi Candro Parulian Sinaga; Endra Ary Prasasty Marpaung; Nera Mayana Br Tarigan; Vinsensius Fereri Purba; Dwicky Aditya
The IJICS (International Journal of Informatics and Computer Science) Vol. 10 No. 2 (2026): July
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/ijics.v10i2.9795

Abstract

Student achievement assessment in educational institutions is often conducted manually by relying on average scores or class rankings, making the results less objective and unable to represent students’ overall academic conditions. This study aims to implement the K-Means Clustering algorithm to classify student achievement using academic performance indicators. The dataset consists of 10 student records with four variables: average score, examination score, attendance percentage, and assignment score. The research stages include data collection, data validation, Min-Max normalization, initial centroid selection, Euclidean Distance calculation, cluster formation, centroid updating, and interpretation of clustering results. The number of clusters was set to K=3, representing high, medium, and low achievement categories. The results show that Cluster C1 consists of 4 students with high achievement, Cluster C2 consists of 3 students with medium achievement, and Cluster C3 consists of 3 students with low achievement. The final centroid values indicate that Cluster C1 has the strongest academic performance, while Cluster C3 requires more intensive academic support. These findings demonstrate that K-Means Clustering can classify student achievement objectively and support data-driven educational decision-making, academic guidance, and targeted learning strategy development.
Decision Support System for Early Detection of Depression Risk Based on Lifestyle Patterns Using the ROC-WASPAS Method Fadlina; Sugi Hartono Sinambela
The IJICS (International Journal of Informatics and Computer Science) Vol. 10 No. 2 (2026): July
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/ijics.v10i2.9799

Abstract

This study develops a decision-support model for prioritizing lifestyle-based depression risk indicators using Rank Order Centroid (ROC) weighting and Weighted Aggregated Sum Product Assessment (WASPAS). The model evaluates five respondents against five criteria: stress level, sleep duration, social interaction, dietary pattern, and physical activity. Criterion priorities were converted into ROC weights, after which the decision matrix was normalized according to cost and benefit orientation. WASPAS then combined the weighted sum and weighted product components using an equal aggregation coefficient. Stress received the largest weight (0.456), followed by sleep duration (0.256), social interaction (0.156), dietary pattern (0.090), and physical activity (0.040). The final preference scores placed A1 first (0.569), followed by A3 (0.557), A2 (0.519), A5 (0.502), and A4 (0.434). These scores represent relative lifestyle profiles within the small study sample; they do not constitute a clinical diagnosis of depression. The model demonstrates a transparent calculation workflow that may support preliminary risk prioritization when its criteria, scales, and weights have been validated by mental-health professionals. Future work should use a larger sample, a validated depression instrument, sensitivity analysis, and external validation before practical screening is considered.
K-Means Clustering for Adaptive Learning Recommendations Based on Student Academic Performance in Vocational Education petti sijabat; Eviyanti Barus; Kristin Sitompul; Sinta Suwanda; Maira Mau Lydia
The IJICS (International Journal of Informatics and Computer Science) Vol. 10 No. 2 (2026): July
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/ijics.v10i2.9806

Abstract

Differences in students' academic abilities create challenges in the learning process because many schools still apply uniform learning strategies without considering individual academic characteristics. This study aims to cluster students based on academic performance using the K-Means Clustering algorithm and to generate adaptive learning recommendations for each cluster. The dataset consisted of academic records from 50 students of SMK Swasta 2 Delima Sari in the 2024/2025 academic year, including assignment scores, Midterm Examination (UTS) scores, Final Examination (UAS) scores, and attendance rates. Data preprocessing was conducted through data cleaning and Min-Max normalization to ensure that each variable contributed proportionally to the clustering process. The K-Means algorithm was implemented with k = 3 to form high, medium, and low academic performance groups. The results show that 12 students (24%) were classified into the High Cluster, 28 students (56%) into the Medium Cluster, and 10 students (20%) into the Low Cluster. The clustering quality was supported by a WCSS value of 2.3714 and a Silhouette Score of 0.6843, indicating reasonably well-separated clusters. Based on the cluster profiles, students in the High Cluster are recommended to receive enrichment activities, students in the Medium Cluster receive regular instruction with periodic feedback, and students in the Low Cluster receive remedial learning and structured mentoring. These findings indicate that K-Means Clustering can support data-driven educational decision-making and provide a practical basis for adaptive learning recommendations in vocational education.
Deep Learning-Based Sentiment Analysis on Social Media Text Using Long Short-Term Memory (LSTM) liska sipayung; Megaria Purba; Bayu Pratama; Sri Yessi Saragih Sumbayak
The IJICS (International Journal of Informatics and Computer Science) Vol. 10 No. 2 (2026): July
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The large volume and informal nature of Indonesian social media text make manual sentiment analysis slow, inconsistent, and difficult to scale. This study evaluates a Long Short-Term Memory (LSTM) model for classifying public comments from X/Twitter into positive, neutral, and negative sentiment. A balanced dataset of 3,000 public Indonesian-language posts collected from January to March 2026 was manually labeled into three equal classes. Duplicate, irrelevant, advertising, and empty posts were removed; the remaining text underwent case folding, noise removal, tokenization, and padding. The data were stratified into 2,400 training and 600 testing instances. The model used a 10,000-word vocabulary, 100-token sequences, a 128-dimensional embedding, 128 LSTM units, dropout of 0.5, and a softmax output layer. On the held-out test set, the model obtained 87.00% accuracy, 86.80% precision, 86.50% recall, and 86.60% F1-score. Positive sentiment produced the strongest class-level performance, whereas neutral comments were more difficult because factual, ambiguous, and mixed expressions provide weaker affective cues. The findings show that LSTM provides a useful baseline for three-class Indonesian social media sentiment classification. However, generalization remains limited by the single-platform, topic-dependent dataset and the absence of repeated or cross-domain evaluation.
Enhancing Product Recommendations Using ALS Matrix Factorization on Retailrocket with Apache Spark Agustina; R. Mahdalena Simanjorang; Bagas Multasyah; Bagus Rivaldi
The IJICS (International Journal of Informatics and Computer Science) Vol. 10 No. 2 (2026): July
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/ijics.v10i2.9845

Abstract

The rapid growth of e-commerce has generated massive volumes of user interaction data, requiring recommendation systems capable of providing accurate and relevant product suggestions. However, conventional recommendation systems still face sparsity issues and low prediction accuracy in large-scale data environments. This study aims to improve the accuracy of big data-based product recommendation systems by implementing Matrix Factorization using the Alternating Least Squares (ALS) algorithm. The research utilized the Retailrocket E-commerce Dataset consisting of 104,287 interaction records, 18,642 users, and 7,831 products with a sparsity level of 96.8%. Implicit interactions, including product views, add-to-cart activities, and purchases, were transformed into weighted preference values to represent user behavior. The model was implemented using Apache Spark MLlib within a distributed computing environment. Model evaluation was conducted using RMSE, MAE, Precision@10, and Recall@10 metrics with a 5-fold cross-validation approach. The experimental results indicate that the optimal configuration was achieved using 50 latent factors, 0.05 regularization, and 20 iterations, producing an RMSE value of 0.836, MAE of 0.689, Precision@10 of 0.861, and Recall@10 of 0.824. These findings demonstrate that ALS-based Matrix Factorization effectively improves recommendation quality while supporting scalability for large-scale data processing in modern e-commerce environments. Keywords: Matrix Factorization, Alternating Least Squares, Recommendation System, Big Data, Collaborative Filtering.
Laundry Performance Analysis Using KPI and Linear Regression Dashboard Dameria Esterlina Br Jabat; Megaria Purba; Putri D Br Sitorus; Eva S Saragih
The IJICS (International Journal of Informatics and Computer Science) Vol. 10 No. 2 (2026): July
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/ijics.v10i2.9859

Abstract

This study develops an integrated approach for measuring and forecasting the revenue performance of a small laundry business using Key Performance Indicators (KPIs), simple linear regression, and a dashboard prototype. The dataset consists of operational records from one Indonesian laundry enterprise covering January–June 2024. Daily records were validated, anonymized, and aggregated into six monthly observations containing transaction volume, customer count, laundry weight, revenue targets, and actual revenue. The KPIs comprised total revenue, month-to-month revenue growth, average revenue per transaction, and target achievement. A chronological holdout was used to evaluate the regression model: January–April formed the training set, whereas May–June formed the test set. The holdout evaluation produced MAE of Rp1,287,500, RMSE of Rp1,319,209, and MAPE of 6.31%. After validation, the model was refitted to all six observations, resulting in Ŷ = 13,216,666.67 + 1,235,714.29X with R² = 0.9198. Total observed revenue was Rp105,250,000; the highest monthly revenue was Rp21,000,000 in June; average revenue per transaction remained Rp50,000; and overall target achievement was 105.25%. The model forecast revenues of Rp21.87 million, Rp23.10 million, and Rp24.34 million for July, August, and September, respectively. The dashboard consolidates the KPI and forecast results into a concise decision-support view. Because the analysis uses one business and only six monthly observations, the forecasts should be interpreted as an exploratory trend estimate rather than a generalizable long-term model.
LSTM-Based Deep Learning Approach for Hoax Detection on Indonesian Social Media Nuraisana nuraisana; R. Mahdalena Simanjorang; Thania Rizky Ristanti; Della Triyani
The IJICS (International Journal of Informatics and Computer Science) Vol. 10 No. 2 (2026): July
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/ijics.v10i2.9899

Abstract

The widespread adoption of social media, particularly Facebook, has drastically accelerated information dissemination while simultaneously amplifying the spread of misinformation. Hoax content poses a serious threat to public trust and social cohesion. This study proposes a hoax classification system based on the Long Short-Term Memory (LSTM) architecture, selected for its inherent ability to model sequential dependencies in textual data. The experimental workflow covered text preprocessing (case folding, tokenization, stopword removal, and stemming), word vectorization using the Keras Tokenizer, and end-to-end LSTM model training on the “Indonesian Fact and Hoax Political News” dataset from Kaggle, totaling 4,502 samples with a balanced 80:20 train-test split. Experimental results demonstrate the proposed model achieves 92.44% accuracy, 92.63% precision, 92.22% recall, and a 92.42% F1-score on the test set. These findings confirm that LSTM effectively captures contextual and sequential linguistic patterns in Indonesian-language content, offering a viable and scalable solution for automated hoax detection on social media platforms.
Development of a Web-Based Nutritious Food Consumption Monitoring System Megaria Purba; Liskedame Yanti Sipayung; Puteri Fajar Addini; Khairunisa; Julia Friska Sahputri
The IJICS (International Journal of Informatics and Computer Science) Vol. 10 No. 2 (2026): July
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/ijics.v10i2.9917

Abstract

School meal programs require reliable records to determine whether meals are distributed and consumed as planned. At the observed school, consumption was recorded manually, creating risks of incomplete entries, delayed recapitulation, and difficulty retrieving historical data. This study developed a web-based nutritious food consumption monitoring system using the Waterfall model. Requirements were identified through observation, stakeholder interviews, and document review, then translated into Unified Modeling Language and database designs. The implemented modules cover role-based login, student records, food menus, daily consumption entries, and period-based reports. Functional verification used predefined black-box scenarios for authentication, data management, consumption recording, reporting, and logout. All seven primary scenarios produced the expected outputs, while two focused authentication cases correctly accepted valid credentials and rejected invalid credentials. These results show that the prototype satisfies the tested functional requirements and centralizes previously fragmented records. However, the 100% functional success rate applies only to the specified test cases and does not establish usability, performance, security, nutritional adequacy, or effectiveness in improving student health. Field deployment with representative users, measured task performance, security assessment, and validation of nutritional indicators is therefore required before operational adoption.

Page 15 of 16 | Total Record : 153

 12   13   14   15   16 

Filter by Year

2017 2026


Reset
Filter By Issues
All Issue Vol. 10 No. 2 (2026): July Vol. 10 No. 1 (2026): Maret Vol. 9 No. 3 (2025): November Vol. 9 No. 2 (2025): July Vol 7, No 1 (2023): March 2023 Vol 6, No 3 (2022): November 2022 Vol 6, No 2 (2022): July 2022 Vol 6, No 1 (2022): March 2022 Vol 5, No 3 (2021): November 2021 Vol 5, No 2 (2021): July 2021 Vol 5, No 1 (2021): March 2021 Vol 4, No 2 (2020): September 2020 Vol 4, No 1 (2020): March 2020 Vol 3, No 2 (2019): September 2019 Vol 3, No 1 (2019): March 2019 Vol 1, No 2 (2017): September 2017 Vol 2, No 1 (2018): March 2018 Vol 1, No 1 (2017): March 2017 More Issue