cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
DIRJA NUR ILHAM
Contact Email
dirja.poltas@gmail.com
Phone
085261233288
Journal Mail Official
dirja.poltas@gmail.com
Editorial Address
Kampus Politeknik Aceh Selatan Jl. Merdeka Komplek Reklamasi Pantai
Location
Kab. aceh selatan,
Aceh
INDONESIA
PERFECT: Journal of Smart Algorithms
Published by Politeknik Aceh Selatan
ISSN : 30640377     EISSN : 30640377     DOI : https://doi.org/10.62671/perfect.v1i1.1
Core Subject : Science, Engineering,
Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Library & Information Science
PERFECT: Journal of Smart Algorithms is an international, Computer Technology, peer-reviewed and open-access journal that provides a platform to produce high-quality original research, Reviews, Letters, and case reports in natural, social, applied, formal sciences, arts, and all other related fields. Our aim is to ameliorate the speedy distribution of new research ideas and results and allow the researchers to create new knowledge, studies, and innovations that will aid as a reference tool for the future. PERFECT is published twice in one year, namely in January and July.
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 24 Documents
 1   2   3 
A Comprehensive Review on Data Science Frameworks for Big Data Analytics Raza, Hassan; Erdenetsogt, Tsendayush; Singh, A; Farooq, Mazhar; Kabeer, Muhammad Mohsin; Aslam, Muhammad Shahrukh
PERFECT: Journal of Smart Algorithms Vol. 3 No. 1 (2026): PERFECT: Journal of Smart Algorithms, Article Research January 2026
Publisher : LEMBAGA KAJIAN PEMBANGUNAN PERTANIAN DAN LINGKUNGAN (LKPPL)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62671/perfect.v3i1.217

Abstract

The importance of big data analytics is now essential in deriving insights in large and complex information in various industries. This review discusses major data science frameworks, such as Apache Hadoop, Spark, Flink, and Storm, their architecture, capabilities, and a relative advantage of processing batches and in real-time. It also presents major challenges that can affect the framework efficiency, including scalability, latency, and heterogeneity of data, security, and the complexity of operational, among others. Lastly, the new trends such as the adoption of AI, cloud-native architecture, real-time streaming, and intelligent automation are discussed to demonstrate the changing environment. This review gives an in-depth insight into the concept of big data frameworks and how they facilitate the achievement of effective analytics.
Machine Learning Driven Decision Making in the Modern Data Era Raza, Hassan; Singh, A; Erdenetsogt, Tsendayush; Kabeer, Muhammad Mohsin; Aslam, Muhammad Shahrukh; Farooq, Mazhar
PERFECT: Journal of Smart Algorithms Vol. 3 No. 1 (2026): PERFECT: Journal of Smart Algorithms, Article Research January 2026
Publisher : LEMBAGA KAJIAN PEMBANGUNAN PERTANIAN DAN LINGKUNGAN (LKPPL)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62671/perfect.v3i1.224

Abstract

The era of modern data has seen an unprecedented increase in the number of data generated, which generates an opportunity as well as a challenge to the decision making. Machine learning (ML) has become the significant solution to work with big and multifaceted data, recognize trends, and provide foresight and change the usual decision-making processes. This review examines the principles, methods and uses of ML-based decision systems in various industries, such as healthcare, finance, retail, transportation and education. It also analyses problems of data quality, bias, transparency, ethical considerations and developments related to explainable and trustworthy AI. Lastly, future trends, human-machine cooperation, and research perspectives are addressed, with a focus on the possibility of the ML to accelerate, more precise and answerable decisions in the world that runs on data.
Optimization of Computer Network Performance through Traffic Management and Bandwidth Allocation Ilham, Dirja Nur; Harahap, Muhammad Khoiruddin; Talib, Muhammed Saat; Budiansyah, Arie; Candra, Rudi Arif
PERFECT: Journal of Smart Algorithms Vol. 3 No. 1 (2026): PERFECT: Journal of Smart Algorithms, Article Research January 2026
Publisher : LEMBAGA KAJIAN PEMBANGUNAN PERTANIAN DAN LINGKUNGAN (LKPPL)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62671/perfect.v3i1.243

Abstract

Computer network performance is highly dependent on effective traffic management and proper bandwidth allocation, especially in network environments with a large number of users and diverse service demands. Uneven bandwidth distribution often leads to degraded service quality, including low throughput, high delay, and increased packet loss. This study aims to analyze and optimize computer network performance through the implementation of traffic management and bandwidth allocation using the Simple Queue and Queue Tree methods on a MikroTik router. An experimental research approach was employed by comparing network performance before and after the application of bandwidth management mechanisms. The evaluation was conducted based on Quality of Service (QoS) parameters, namely throughput, delay, and packet loss, in accordance with the TIPHON standard. The experimental results indicate a significant improvement in network performance after the implementation of Simple Queue and Queue Tree. Throughput increased substantially, while delay and packet loss were considerably reduced, resulting in improved service quality categories. The Simple Queue method effectively ensured fair bandwidth distribution among users by limiting per-user bandwidth usage, whereas the Queue Tree method enhanced performance by prioritizing network traffic based on service types. The combination of these methods successfully minimized bandwidth monopolization, reduced network congestion, and improved overall network stability. Therefore, the implementation of Simple Queue and Queue Tree proves to be an effective solution for optimizing bandwidth utilization and enhancing Quality of Service in computer networks with high user density and heterogeneous traffic characteristics.
LightEmoNet: Lightweight Deep Learning for Facial Emotion Recognition Kamber, Ali Nadhim; Alkaabi, Hussein Alaa
PERFECT: Journal of Smart Algorithms Vol. 3 No. 1 (2026): PERFECT: Journal of Smart Algorithms, Article Research January 2026
Publisher : LEMBAGA KAJIAN PEMBANGUNAN PERTANIAN DAN LINGKUNGAN (LKPPL)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62671/perfect.v3i1.273

Abstract

Facial emotion recognition (FER) is a critical component of human-computer interaction, affective computing, and intelligent surveillance systems. Existing deep learning approaches, while achieving high accuracy, are often computationally expensive and unsuitable for deployment on resource-constrained or real-time systems. In this paper, we present LightEmoNet, a lightweight Convolutional Neural Network (CNN) architecture specifically designed for efficient and accurate facial emotion recognition. Our model is trained on the FER2013 benchmark dataset, which contains 35,887 grayscale images distributed across seven emotion classes: Happy, Neutral, Sad, Fear, Angry, Surprise, and Disgust. To address the inherent class imbalance within the dataset, we employ a dual strategy combining class-weighted loss penalization with targeted data augmentation applied selectively to underrepresented categories. The proposed architecture totals approximately 2.1 million trainable parameters and occupies only 8.3 MB on disk, making it deployable on edge and embedded platforms without GPU acceleration. Experimental results demonstrate that LightEmoNet achieves a training accuracy of 91.0% and a validation accuracy of 88.5% on the FER2013 test split, with an average inference latency of 4.2 ms per image on a standard CPU. The model exhibits robust performance across all seven emotion classes while maintaining a compact footprint suitable for real-time inference. These findings confirm that lightweight CNNs, when paired with principled augmentation strategies, can achieve competitive performance without the overhead of large-scale deep models.

Page 3 of 3 | Total Record : 24

 1   2   3 

Filter by Year

2024 2026


Reset
Filter By Issues
All Issue Vol. 3 No. 1 (2026): PERFECT: Journal of Smart Algorithms, Article Research January 2026 Vol. 2 No. 2 (2025): PERFECT: Journal of Smart Algorithms, Article Research July 2025 Vol. 2 No. 1 (2025): PERFECT: Journal of Smart Algorithms, Article Research January 2025 Vol. 1 No. 2 (2024): PERFECT: Journal of Smart Algorithms, Article Research July 2024 Vol. 1 No. 1 (2024): PERFECT: Journal of Smart Algorithms, Article Research January 2024