Agung Trisetyarso’s research while affiliated with Multimedia Nusantara University and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (107)


Figure 1. Stages of predicting the expected fault output by algorithms
Multi-algorithm for predicting the level accuracy of fault output in software
  • Article
  • Full-text available

December 2024

·

17 Reads

Zulkifli Zulkifli

·

·

Agung Trisetyarso

·

Software fault output refers to software errors found by testers during the software testing process. Software testing is an overly critical stage in software development; hence, a software testing model is required to systematically classify software errors. For the process of classifying software fault output, accuracy measurements are needed to predict manual fault outputs compared to algorithm-generated fault outputs. Algorithmic methods can be used to measure the accuracy of fault output. The main objective of this research was to compare different algorithms for predicting the accuracy of fault output on a dataset derived from past software testing. The findings indicate that the neural network algorithm outperforms SVM, MLP, RF, and MNB algorithms, achieving 98% accuracy when using 10 software testing variables (function, interface, structure, performance, requirement, documentation, positive, negative, basis path, and times) to predict expected fault outputs.

Download





FIGURE 1. Office31 sample dataset for each domain: Amazon, DSLR, Webcam
FIGURE 2. Architecture of Adversarial Multitask Learning (AML). This method utilizing adversarial learning and multitask learning to minimize the domain discripancies through domain adapter
FIGURE 3 Instance of domain adaption process by utilizing adversarial multitask learning. It shows the loss value over the epochs of the training. The convergence of the loss function indicates that adversarial multitask learning has successfully learned to minimize domain discrepancies and optimize the model parameters, resulting in enhanced domain adaptation performance.
The best experiment results for each scenario from grid search experiment using combination of different data source and target with different weights on its loss values.
Adversarial Multitask Learning for Domain Adaptation through Domain Adapter

January 2024

·

16 Reads

IEEE Access

This study presents a technique called Adversarial Multitask Learning (AML) to enhance the effectiveness of domain adaptation methods in practical applications, which are currently highly sought after. The proposed approach addresses the challenges posed by domain shift by effectively managing multiple interconnected tasks through the principles of adversarial training. Utilizing the widely recognized Office31 dataset, we assess the efficacy of our model across different domains. Our approach employs a multitask learning paradigm, focusing on adapting to the target domain while leveraging shared feature representations through classification tasks. This strategy ensures that primary and auxiliary tasks incorporate domain-invariant properties, allowing for robust adaptation to varying domains. The results reveal significant improvements in adaptation performance when compared to conventional domain adaptation techniques. For instance, the accuracy of our AML model in adapting from the webcam domain to the dslr domain reached 88.54%, surpassing 86.46% for models without adversarial training and 78.91% for those lacking categorical training. Furthermore, we conduct a comprehensive examination of how different hyperparameters influence model performance, enhancing our understanding of the fundamental mechanisms underlying Adversarial Multitask Learning for domain adaptability. Overall, this paper contributes significantly to the field of domain adaptation by introducing the AML framework, underscoring the importance of further exploration of multitask learning paradigms and adversarial training to improve domain adaptation in real-world scenarios.






Citations (60)


... By combining these software tools, researchers enhanced the analysis process. they gained more profound insights into the respective fields of study, showcasing the importance of utilizing multiple software programs for comprehensive research analysis (Fatima & Quamer, 2023;nurhidayah et al., 2024;soegoto et al., 2023) this paper uses a separate exploration of literature analysis methodologies: evaluation of performance, scientific mapping and analysis of networks (Donthu et al., 2021). the research method is summarized in Figure 1. ...

Reference:

Mapping research landscape of emerging technology in the accounting field: a bibliometric analysis
Analysis of Enterprise Architecture Research Trends for Higher Education Institutions Using Systematic Literature Review and Vos Viewer
  • Citing Conference Paper
  • October 2023

... Moving to practical implementations, Yulianti et al. (2023) improve ensemble classifiers using a hybrid quantum annealing method, and Li et al. (2023) introduce an innovative quantum approach to k-fold cross-validation, simplifying classification tasks. These developments highlight the impact of quantum computing in improving traditional machine learning techniques. ...

A Hybrid Quantum Annealing Method for Generating Ensemble Classifiers
  • Citing Article
  • November 2023

Journal of King Saud University - Computer and Information Sciences

... The authors compared four distance metrics, including the Euclidean, Manhattan, Chebyshev, and Canberra distances, in the clustering method for ensemble generation, and evaluated them based on accuracy [8]. The Canberra distance consistently achieved better results than the others, which motivates our work to explore its application in classification tasks. ...

Comparison of Distance Metrics for Generating Cluster-based Ensemble Learning
  • Citing Conference Paper
  • June 2023

... Both annotation and parameter tweaking can be automated. The YOLOv3-ResNet18 system is developed in [43] for 4 fish species detection and classification on the private dataset. Comparative analysis with SSD-VGG and Huawei ExeML models demonstrates the superior performance of YOLOv3-ResNet18, achieving an accuracy of 98.45%. ...

Fish Classification System Using YOLOv3-ResNet18 Model for Mobile Phones

CommIT (Communication and Information Technology) Journal

·

·

Agung Trisetyarso

·

[...]

·

... Subsequently, the research was extended by creating an integration-based model (I-BM) framework for MBT. The I-BM framework employed 8 variables: function, interface, structure, performance, requirement, documentation, positive, and negative to predict the accuracy level of its fault output using machine learning approaches (Zulkifli et al., 2023). Furthermore, Ali et al. (2023) conducted research on defect prediction in software using machine learning. ...

Software Testing Integration-Based Model (I-BM) Framework for Recognizing Measure Fault Output Accuracy Using Machine Learning Approach
  • Citing Article
  • May 2023

International Journal of Software Engineering and Knowledge Engineering

... Más del 70% de los clientes demandan productos y servicios digitales innovadores, mientras que la industria aún no está preparada para esta era. Esto exige repensar los indicadores clave de desempeño (KPIs), ya que, para triunfar en este nuevo entorno, los bancos deben centrarse en las opiniones de los usuarios, la imagen y la estructura financiera de la tecnología [3]. ...

Capability Development to Generate Business Value Through Customer-centric Analytics in the Banking Industry: A Systematic Review

Journal of System and Management Sciences

... For example, in a comprehensive review, Afriliana & Ramadhan (2022) discussed the trends of using robotic process automation technology for digital automation. Similarly, Marcel et al. (2023) proposed an extensive framework for the digital transformation of organizations for their business sustainability and to gain a competitive advantage in the market. Similarly, Adityawan et al. (2023) considered the case study of using digitization in the Islamic banking sector in Indonesia to get a competitive advantage. ...

Digital Transformation Adoption: An Extended Step-by-Step Framework

Journal of System and Management Sciences

... Numerous research works have employed algorithms to forecast the precision of defects associated with the expected software fault output. One notable study undertaken by Zulkifli et al., (2022), developed a modelbased test (MBT) using 5 variables, including performance, initialization, data structures, incorrect or missing functions, and external database access, with the neural network algorithm. Subsequently, the research was extended by creating an integration-based model (I-BM) framework for MBT. ...

Software Testing Model by Measuring the Level of Accuracy Fault Output Using Neural Network Algorithm
  • Citing Conference Paper
  • November 2022

... In the Metaverse, individuals can engage in virtual meetings where body language and eye contact can be used to convey emotions and intentions, enhancing communication and collaboration. By leveraging the immersive nature of the Metaverse, telecommuting experiences can be significantly improved, allowing coworkers to connect and collaborate seamlessly from diverse perspectives, regardless of physical distance [28][29][30]. ...

Acceptance of augmented reality in video conference based learning during COVID-19 pandemic in higher education

Bulletin of Electrical Engineering and Informatics

... Existen diversos softwares en el mercado que contemplan la inteligencia de negocios para extraer, transformar y cargar los datos (proceso ETL) permitiendo visualizar y analizar la información tales como: Microsoft Power BI, Tableau, Oracle Business Intelligence, Pentaho BI, IBM Cognos, Sap Business Intelligence, Looker, Qlik Sense y Sisense (Yanfi et al., 2022;García Pérez, 2020). Es posible crear dashboards (tableros de control) utilizando estos softwares, los cuales son herramientas visuales que permiten monitorear y analizar métricas clave de manera clara y en tiempo real, facilitando aún más la toma de decisiones empresariales que no solo facilitan la visualización de información, sino que también permiten tomar medidas correctivas y preventivas de manera oportuna. ...

Measuring Student’s Satisfaction and Loyalty on Microsoft Power BI Using System Usability Scale and Net Promoter Score for the Case of Students at Bina Nusantara University