cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Syaifudin
Contact Email
jurnal_intelmatics@trisakti.ac.id
Phone
+628129513950
Journal Mail Official
jurnal_intelmatics@trisakti.ac.id
Editorial Address
Building E, floor 4, Department of Informatics Engineering, Universitas Trisakti
Location
Kota adm. jakarta barat,
Dki jakarta
INDONESIA
Intelmatics
Published by Universitas Trisakti
ISSN : -     EISSN : 27758850     DOI : https://doi.org/10.25105/itm
Core Subject : Science,
Computer Science & IT
The IntelMatics Journal is a scientific journal published by the department of informatics engineering at Trisakti University. The purpose and objective of the publication of the IntelMatics journal are as a means of dissemination of international standard science in the field of software engineering, information security, and business analysis in the scope of data intelligence and visualization. Journal will be published every sixth month
Arjuna Subject : Teknik (semua kategori) - Teknik (Lain-Lain)
Articles 80 Documents
 4   5   6   7   8 
Decision Support System for Competition Assessment Using the Analytical Hierarchy Process and Multi-Attribute Utility Theory Putri Afiifah, Nabiilah; Hidayati, Anita
Intelmatics Vol. 5 No. 2 (2025): July-December
Publisher : Penerbitan Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/v5i2.24060

Abstract

The advancement of information technology has driven organizations to undergo digital transformation, including in the implementation of competitions. The Informatics and Computer Engineering Student Association (HIMATIK) of Politeknik Negeri Jakarta still uses spreadsheets for registration and judging processes, which are prone to human error, inefficiency, and lack of transparency. This study aims to design and develop a web-based Decision Support System (DSS) using the Analytical Hierarchy Process (AHP) and Multi Attribute Utility Theory (MAUT) methods to improve the quality of judging. The system was developed using Laravel, PHP, MySQL, and front-end technologies such as HTML, CSS, and JavaScript. The user interface was designed using Figma, and the development process followed the Waterfall methodology. AHP was used to determine the weight of the criteria, while MAUT was applied for score calculation and participant ranking. The research approach included qualitative methods and quantitative methods. Black box testing results showed that the system functioned 100% as expected. The System Usability Scale (SUS) score of 89.19 (Excellent category) and a Net Promoter Score (NPS) of 67.5% (Very Good category) indicate a high level of user satisfaction. This system is expected to enhance efficiency, accuracy, and transparency in the judging process and support the digitalization of future HIMATIK activities.
Development of An Employee Training Information System Using The Deductive Training Model Fauziyyah, Luthfiyyah Zharifa
Intelmatics Vol. 5 No. 2 (2025): July-December
Publisher : Penerbitan Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/v5i2.24074

Abstract

The Financial Transaction Reports and Analysis Center (PPATK) provides training as a mandatory requirement to improve the competencies of its employees. However, the training registration process has not been running optimally, as it still involves multiple impractical stages and makes it difficult to monitor compliance with training requirements based on existing policies. To address this issue, this study aims to develop an employee training information system that aligns with each employee’s competencies and supports supervision of training activities across the organization to ensure compliance with established policies, thereby preventing competency and performance gaps among employees. The system development adopts a deductive training model tailored to the characteristics and needs of employees based on their respective work units. The research applies the Rapid Application Development (RAD) method, using PHP programming language with the Laravel framework and MySQL database. Data collection was carried out through observation, interviews, and literature study to understand user requirements. The SWOT analysis was conducted to evaluate the existing system. Thereafter, the newly developed system was tested using the black box method to rigorously assess its functionality. The test results indicate that all features, including training registration, duration recording, and total training hours monitoring, function as designed without errors. Based on the development and testing results, this system can be recommended as a digital, integrated, and needs-based alternative to support the management and supervision of employee training within the organization
UX Comparison of Physical Traces, Samsung Health, Strava with EUCS Method Rini, Alefia; Shofiati, Ratna; Zuhdi, Ahmad
Intelmatics Vol. 6 No. 1 (2026): January-June (In Progress)
Publisher : Penerbitan Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/v6i1.23929

Abstract

The development of digital technology has encouraged people to be more concerned about their health through the use of physical activity monitoring applications. This study aims to compare user experiences of three physical activity tracking applications, namely Physical Traces (developed by researchers), Samsung Health, and Strava, using the End User Computing Satisfaction (EUCS) approach. EUCS assesses user satisfaction based on five dimensions: content, accuracy, format, ease of use, and timeliness. Data were collected through a 4-point Likert scale-based questionnaire distributed to 107 respondents. The results showed that all questionnaire items were valid and reliable. The Samsung Health application excelled in the content and timeliness dimensions, Strava excelled in accuracy, while Physical Traces had the highest scores in ease of use and format. However, a statistically significant difference was found only in the accuracy dimension (p = 0.0016). These findings provide an overview of the strengths and weaknesses of each application, as well as input for improving the Physical Traces application in the future.
Web-Based System for Managing Garage Service And Spare Parts Inventory Muhammad Danil Hidayat; Pratiwi, Dian; Agus Salim
Intelmatics Vol. 6 No. 1 (2026): January-June (In Progress)
Publisher : Penerbitan Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/v6i1.24712

Abstract

LangSpeed Motor Garage faces operational challenges, including data inaccuracies and slow manual processes in service and spare parts inventory management, which affect efficiency. This research aims to design, build, and test an integrated web-based management information system to address these problems. The development method used is the Waterfall model, which includes the stages of requirements analysis, system design using UML, code implementation, and system testing. This system was built as a single-page application (SPA) with a modern technology stack, namely ReactJS on the front-end, the serverless Convex backend platform for a real-time database, and Tailwind CSS for interface design. The result of this research is a functional web application that features service management, inventory, and user authentication modules. Based on testing using the Black Box Testing method, all functionality test scenarios were successful with a "Passed" status, which validates that the system runs according to the requirements specifications. The developed information system has proven to be an effective solution for improving operational efficiency and data accuracy at LangSpeed Motor Garage.
Comparative Analysis of LightFM and Neural Collaborative Filtering for Anime Recommendation Systems Farras, Daffa Haidar; Rahmatulloh, Alam
Intelmatics Vol. 6 No. 1 (2026): January-June (In Progress)
Publisher : Penerbitan Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/v6i1.24815

Abstract

The rapid growth of digital entertainment content such as movies and anime poses challenges in providing relevant viewing recommendations to users. Recommendation systems are a crucial solution to improve the user experience when exploring thousands of titles across various platforms. This study develops and compares two main approaches to recommendation systems: LightFM and the Collaborative Method . Classic Filtering (CF), as well as Neural Collaborative Deep- based Filtering (NCF) learning . Evaluation was conducted using Precision@K and Recall@K metrics . The test results showed that NCF was able to provide more relevant recommendations, with a Precision@5 value of 0.7983, much higher than LightFM which only reached 0.1721. Although LightFM showed a high AUC value (0.9134), its performance in generating Top-K recommendations was still low. Thus, it can be concluded that modern neural network -based approaches such as NCF are more effective than classical methods in the context of anime recommendation systems.
Exploratory Data Analysis of Agricultural Commodity Prices Across Districts and Cities in East Java Tamam, Moh. Badri; Araujo Bernardo, Januario Freitas
Intelmatics Vol. 6 No. 1 (2026): January-June (In Progress)
Publisher : Penerbitan Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/v6i1.25460

Abstract

Descriptive statistics and visualization revealed significant variations in price distributions, with medium rice, premium rice, and dry milled paddy being the predominant commodities. Monthly trend analysis indicated a consistent upward movement in prices from 2020 to mid-2025. Regional comparisons specifically highlighted that, among all districts studied, Bondowoso had the highest average rice prices, Magetan consistently showed the lowest prices, and Malang City exhibited the greatest price volatility. The Random Forest Classifier achieved 73% accuracy in distinguishing districts based on commodity price patterns. Feature importance analysis singled out medium rice and premium rice as the most influential variables driving classification. These key findings clarify the specific pricing trends, regional disparities, and significant predictors identified in the study.
Development of a Promt Chatbot Application based on Rasa Riansyah, Rafli; Rochman, Abdul; Budi Santoso, Gatot
Intelmatics Vol. 6 No. 1 (2026): January-June (In Progress)
Publisher : Penerbitan Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/v6i1.26031

Abstract

Abstract: The development of artificial intelligence-based chatbots has increased the need for systems that can support interactive services efficiently. The Rasa framework is widely used in chatbot development due to its flexibility, but the prompt design process is still done manually by editing configuration files, making it difficult for non-technical users. This study aims to develop a web-based application that facilitates the design and management of Rasa-based chatbot prompts without the need for direct code editing. The system development method used is Waterfall, which includes the stages of requirements analysis, design, implementation, testing, deployment, and maintenance. The application was developed using the Laravel framework and integrated with the Rasa chatbot structure. System testing was conducted using the Black Box Testing method to ensure that the application functions as required. The results of the study show that the application simplifies the management of chatbot intents, training data, entities, and responses, thereby increasing efficiency and ease of use for users in designing Rasa-based chatbots.
Implementation of Odoo Enterprise Resource Planing (ERP) in the Jewelry Repair Process Using the Repair Module at XYZ Store Ubaidillah, Fariz; Syaifudin; Siswanto, Teddy
Intelmatics Vol. 6 No. 1 (2026): January-June (In Progress)
Publisher : Penerbitan Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/v6i1.26150

Abstract

The jewelry repair sector continues to develop alongside increasing consumer demand for maintenance, resizing, and restoration services. However, many small and medium jewelry repair businesses still rely on manual administrative processes, resulting in inaccurate repair records, limited visibility of repair status, and delays in service completion. This study focuses on the implementation of an Odoo-based Enterprise Resource Planning (ERP) system by utilizing the Repair module as the core component to manage jewelry repair processes in a structured and integrated manner. The implementation is supported by related modules such as Inventory and Accounting to ensure material tracking and financial recording during the repair lifecycle. The Accelerated SAP (ASAP) methodology is applied, covering Project Preparation, Business Blueprint, Realization, and Final Preparation stages. The results demonstrate improvements in repair process efficiency, data accuracy, transparency of repair status, and real-time . This study concludes that the Odoo Repair module provides an effective solution for managing jewelry repair operations and recommends future research on automation of repair scheduling and customer notification systems.
Semantic Similarity Implementation for The Circle Application Support Group using Tensorflow Language Models and Scalable Nearest Neighbor Rizkani, Nadiya; Mardianto, Is; Ariwibowo, Anung Barlianto; Gunawan, M
Intelmatics Vol. 6 No. 1 (2026): January-June (In Progress)
Publisher : Penerbitan Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/v6i1.26185

Abstract

Mental health is a condition where individuals are free from all forms of symptoms of mental disorders. This project originated from our desire to find solutions to mental health problems in ourselves. It is often difficult to recognize, express, and tell people we know what we feel. Circle is a technology-based mental health application that supports self-help with meditation, support groups, and online counseling with professional psychologists. To make it easier for users to find chat rooms with similar mental health backgrounds in the Support Group, the Support Group search uses full-text matching based on the user-entered subject.
Evaluation of CNN, BiLSTM, and Hybrid Architectures for Sentiment Analysis of Indonesian X (Twitter) Data Using IndoBERT Embeddings Gintings, Theo Christ Maximos; Ir. Agung Sediyono, MT, Ph.D; Najih, Muhammad
Intelmatics Vol. 6 No. 1 (2026): January-June (In Progress)
Publisher : Penerbitan Universitas Trisakti

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25105/v6i1.26232

Abstract

The rapid growth of social media, particularly X (Twitter), has generated massive volumes of informal and highly contextual opinion data, making sentiment analysis a challenging task. This study aims to systematically evaluate the performance of Convolutional Neural Network (CNN), Bidirectional Long Short-Term Memory (BiLSTM), and two hybrid architectures (CNN BiLSTM and BiLSTM–CNN) in classifying Indonesian-language Twitter sentiment related to the Makan Bergizi Gratis (MBG) policy using frozen IndoBERT embeddings. A quantitative comparative experimental design with stratified 10 fold cross-validation was employed on a dataset of 1,569 tweets that underwent case folding, cleansing, and normalization. Model performance was assessed using accuracy, precision, recall, F1score, ROC-AUC, and computational efficiency (training and inference time), while latent feature quality was analyzed through t-SNE visualization. The results show that hybrid architectures provide more stable and competitive performance than single models. CNN achieved the fastest computation, whereas BiLSTM and hybrid models were superior in capturing sequential context. tSNE visualizations further indicate clearer class separation for hybrid models. These findings confirm that under a uniform IndoBERT embedding, hybrid CNN–BiLSTM and BiLSTM CNN offer the best trade-off between accuracy and efficiency for Indonesian Twitter sentiment analysis.

Page 8 of 8 | Total Record : 80

 4   5   6   7   8 

Filter by Year

2021 2026


Reset
Filter By Issues
All Issue Vol. 6 No. 1 (2026): January-June (In Progress) Vol. 5 No. 2 (2025): July-December Vol. 5 No. 1 (2025): January-June Vol. 4 No. 2 (2024): Juli-Desember Vol. 4 No. 1 (2024): Januari-Juni Vol. 3 No. 2 (2023): Juli-Desember Vol. 3 No. 1 (2023): Januari-Juni Vol. 2 No. 2 (2022): Juli-Desember Vol. 2 No. 1 (2022): Januari-Juni Vol. 1 No. 2 (2021): Juli - Desember Vol. 1 No. 1 (2021): January More Issue