cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Sopiyan Dalis
Contact Email
sopiyan.spd@bsi.ac.id
Phone
+6281380852868
Journal Mail Official
jurnal.paradigma@bsi.a.cid
Editorial Address
Jl. Kramat Raya No.98, Kwitang, Kec. Senen, Kota Jakarta Pusat, DKI Jakarta 10450
Location
Kota adm. jakarta barat,
Dki jakarta
INDONESIA
Paradigma
Published by Universitas Bina Sarana Informatika
ISSN : 14105063     EISSN : 25793500     DOI : http://dx.doi.org/10.31294/paradigma
Core Subject : Science,
Computer Science & IT
The Paradigma Journal is intended as a medium for scientific studies of research, thought and analysis-critical issues on Computer Science, Information Systems, and Information Technology, both nationally and internationally. The scientific article refers to theoretical reviews and empirical studies of related sciences, which can be accounted for and disseminated nationally and internationally. Paradigma Journal accepts scientific articles research at Expert Systems, Information Systems, Web Programming, Mobile Programming, Games Programming, Data Mining, and Decision Support Systems.
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 5 Documents
 1 
Search results for , issue "Vol. 27 No. 2 (2025): September 2025 Period" : 5 Documents clear
Fine-Tuned Autoencoder Neural Network for Anomaly Detection in Accounting Transactions Nur Alamsyah; Budiman, Budiman; Rahmani, Hani Fitria; Erpurini, Wala
Paradigma - Jurnal Komputer dan Informatika Vol. 27 No. 2 (2025): September 2025 Period
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/p.v27i2.8697

Abstract

Anomaly detection in accounting transactions plays a crucial role in identifying irregularities that may signal fraud, errors, or unusual financial behavior. Traditional rule-based and statistical methods often struggle to detect complex and hidden patterns in large-scale financial datasets. This paper presents a fine-tuned Autoencoder Neural Network for detecting anomalies in structured accounting records. The model processes feature such as date, account type, debit, credit, transaction category, and payment method. Preprocessing includes handling missing values, encoding categorical data, and extracting temporal features. The Autoencoder architecture was optimized using multiple hidden layers and dropout regularization to prevent overfitting. Reconstruction errors were used to determine anomaly scores, with a dynamic threshold set at the 98th percentile. Experimental results show that the model accurately distinguishes normal and anomalous transactions, identifying 2,000 outliers from a total of 100,000 records. Additional analysis indicates that anomalies often occur during weekends or holidays and involve unusual payment methods. These findings demonstrate the potential of the fine-tuned Autoencoder as a scalable and intelligent anomaly detection framework to support auditors and financial analysts in proactive fraud prevention.
Modification of Additive Ratio Assessment Method through Distance-Based Weighting Approach for Optimizing Assessment Accuracy Gunawan, Rakhmat Dedi; Arshad, Muhammad Waqas; Wahyudi, Agung Deni; Suryono, Ryan Randy; Widodo, Tri; Ulum, Faruk
Paradigma - Jurnal Komputer dan Informatika Vol. 27 No. 2 (2025): September 2025 Period
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/p.v27i2.8810

Abstract

The Additive Ratio Assessment (ARAS) method is one of the approaches in multi-criteria decision making (MCDM) used to determine the best alternative based on a number of predetermined criteria. The drawback of this method is its heavy reliance on the accuracy of the criterion weighting determination; non-objective weights can lead to biased results. This study aims to improve the accuracy of ranking in multicriteria decision-making through the modification of the ARAS method with a distance-based weighting approach called ARAS-D. The ARAS method, known for its simplicity in calculation, was modified to be more responsive to the distribution of alternative data on each criterion. This distance-based weighting approach objectively determines the weight of the criteria based on variations in data performance, thereby reducing subjectivity in the weighting process. A case study was conducted on the selection of a new store location with six main criteria: rental cost, building area, accessibility, consumer traffic, parking availability, and infrastructure. The results of the evaluation show that the ARAS-D method is able to produce more precise ratings than the standard approach. Store locations with the highest utility value are recommended as the best choice, proving the effectiveness of the method in supporting strategic decisions. The results of the New Store Location 5 alternative rating obtained the highest score with a value of 0.9083, indicating that this location is the most optimal choice overall. This is followed by New Store Location 3 with a value of 0.8617 and New Store Location 1 with a value of 0.8415, which also shows excellent performance against the criteria that have been set. This research contributes to the development of more adaptive and data-based decision-making methods.
K-Means++ and TF-IDF for Grouping Library Books by Topic Pamput, Jessicha Putrianingsih; Muthmainnah, Aindri Rizky; Risal, Andi Akram Nur; Surianto, Dewi Fatmarani
Paradigma - Jurnal Komputer dan Informatika Vol. 27 No. 2 (2025): September 2025 Period
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/p.v27i2.8272

Abstract

The grouping of library materials in the Department of Informatics and Computer Engineering (JTIK) at Universitas Negeri Makassar (UNM) is still conducted using a conventional system that relies on predefined categories and librarian intuition. This approach often leads to inconsistencies in book categorization, making it difficult for users to find relevant references efficiently. To address this issue, this research applies the K-Means++ clustering method, which optimizes centroid initialization for more accurate cluster formation. Books are grouped based on the TF-IDF weighting matrix, resulting in six distinct clusters characterized by unique centroid values. Analysis of the top 10 words per cluster highlights dominant topics within each group. The clustering quality was evaluated using the Silhouette Coefficient, with the highest value of 0.04299, indicating a well-separated cluster structure. These findings demonstrate that K-Means++ effectively organizes books based on word similarity, enhancing library material management and improving information retrieval in the JTIK library.
Implementation of the Standard Deviation Multi-Objective Optimization by Ratio Analysis Method in Warehouse Staff Recruitment Selection Putra, Farhan Nopransyah; Priandika, Adhie Thyo
Paradigma - Jurnal Komputer dan Informatika Vol. 27 No. 2 (2025): September 2025 Period
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/p.v27i2.8373

Abstract

The warehouse staff selection process has a crucial role in ensuring optimal operational efficiency and logistics management. A selection approach that considers aspects of technical skills, work experience, and compatibility with the organization's culture is essential in ensuring the efficiency and effectiveness of logistics management. The labor selection process, including in the context of warehouse staff recruitment, often faces challenges due to subjectivity in decision-making. The implementation of the SD-MOORA method is the main goal in this study in the process of accepting warehouse staff to improve the objectivity and accuracy of candidate selection, the results of this study are expected to contribute to improving the efficiency of the labor selection process and support data-based decision-making in human resource management. The data used in this study consists of 8 candidates and 6 criteria in the selection of warehouse staff admission. The final outcome of optimizing the SD-MOORA method for ranking warehouse staff admissions shows that GT secured the top rank with a value of 0.3827, indicating it is the most suitable candidate according to the selection criteria. AN followed in second place with a score of 0.3752, and BD placed third with a score of 0.3579. This study significantly contributes to advancing the development of decision support systems for warehouse staff selection by applying the SD-MOORA method. By objectively considering the weighting of criteria using standard deviations, this approach enhances both the accuracy and transparency of candidate rankings.
Support Vector Machine with FastText Word Embedding for Hate Speech Aspect Categorization Mardiana, Aida Milati; Rozi, Imam Fahrur; Arianto, Rudy
Paradigma - Jurnal Komputer dan Informatika Vol. 27 No. 2 (2025): September 2025 Period
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/p.v27i2.5127

Abstract

Freedom of expression on Twitter often leads to issues such as hate speech, which may include provocation, incitement, or insults based on race, religion, gender, and other aspects. To address this issue, machine learning techniques can be applied to automatically classify hate speech. Therefore, this study aims to implement a machine learning–based approach for automatic hate speech aspect classification and to evaluate the accuracy of the obtained results. Support Vector Machine is used as the classifier method, with FastText as the word embedding method in the categorization process of hate speech aspects. The categorized aspects include abusive, individual, group, religion, race, physical, gender and other. The dataset used in this research is a collection of Indonesian tweets from Kaggle, which have been classified into each aspect. This study also tested combinations of preprocessing methods, namely filtering with stemming and the FastText pre-trained model. From the test results of the application of the Support Vector Machine method with FastText word embedding, with parameters C value = 1.0, gamma = 1.0 and RBF kernel and the ratio between training data and testing data is 90:10, the best results were obtained accuracy 98%, precision 98%, recall 98% and F1-Score 97% on Physical and Gender aspects. In addition, this study also tested if it did not use fasttext word embedding and the accuracy results showed 84%, precision 74%, recall 86% and F1 Score 79% in the abbusive aspect.

Page 1 of 1 | Total Record : 5

 1 

Filter by Year

2025 2025


Reset
Filter By Issues
All Issue Vol. 27 No. 2 (2025): September 2025 Period Vol. 27 No. 1 (2025): March 2025 Period Vol. 26 No. 2 (2024): September 2024 Period Vol. 26 No. 1 (2024): March 2024 Period Vol. 25 No. 2 (2023): September 2023 Period Vol. 25 No. 1 (2023): March 2023 Period Vol. 24 No. 2 (2022): September 2022 Period Vol. 24 No. 1 (2022): Periode Maret 2022