Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal Venn: Journal of Sustainable Innovation on Education, Mathematics and Natural Sciences Journal of Mathematics, Computation and Statistics (JMATHCOS)
Afif Budi Andy B
Universitas Sulawesi Barat
Education Environmental Science Mathematics Physics Other

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Perbandingan Analisis Sentimen Ulasan Produk pada Platform E-Commerce Menggunakan Algoritma Naïve Bayes dan Random Forest Afif Budi Andy B; Kusnaeni Kusnaeni; Irwan Usman; Muhammad Hidayatullah; Muh. Rifandi
Venn: Journal of Sustainable Innovation on Education, Mathematics and Natural Sciences Vol. 5 No. 3 (2026): Riset Matematika dan Pendidikan Matematika
Publisher : Pusat Studi Bahasa dan Publikasi Ilmiah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.53696/venn.v5i3.458

Abstract

Sentiment analysis has become increasingly important in e-commerce because product reviews influence consumer purchasing decisions and provide feedback for sellers to evaluate product quality and improve services. The large number of online reviews on e-commerce platforms makes manual analysis inefficient and time-consuming, thereby requiring automated sentiment classification methods that are accurate and computationally efficient. This study aims to compare the performance of the Multinomial Naïve Bayes and Random Forest algorithms in classifying sentiment in Tokopedia product reviews using the PRDECT-ID dataset, which consists of 5,400 Indonesian-language reviews. The research methodology involved several preprocessing stages, including case folding, cleaning, normalization, tokenization, stopword removal, and stemming using the Sastrawi library, followed by feature extraction using the TF-IDF method. The dataset was divided using a stratified random split approach with 80% training data and 20% testing data, and the models were evaluated using accuracy, precision, recall, F1-score, and ROC-AUC metrics. The results indicate that Multinomial Naïve Bayes outperformed Random Forest, achieving an accuracy of 93.59%, precision of 91.82%, recall of 94.65%, F1-score of 93.21%, and ROC-AUC of 0.9813. In comparison, Random Forest achieved an accuracy of 90.35%, precision of 85.63%, recall of 93.67%, F1-score of 89.47%, and ROC-AUC of 0.9635. In addition to its superior classification performance, Multinomial Naïve Bayes also demonstrated greater computational efficiency with significantly faster training time. These findings suggest that Multinomial Naïve Bayes is a more effective approach for sentiment classification of Indonesian-language e-commerce product reviews.
Deepfake Image Classification Using ResNet50 Feature Extraction and XGBoost Learning Model Kusnaeni Kusnaeni; Ika Reskiana Adriani; Mega Sartika Hafid; Afif Budi Andy B; Muhammad Edy Rizal
Journal of Mathematics, Computations and Statistics Vol. 8 No. 2 (2025): Volume 08 Nomor 02 (Oktober 2025)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/jmathcos.v8i2.8387

Abstract

Deepfake is an artificial intelligence-based media manipulation technology that realistically fabricates a person's face, voice, and movements in both video and audio formats. The increasing use of deepfakes in the creation of various forms of deceptive content, including pornography, fake news, and fraud, has led to an urgent need for effective detection methods. One of the main challenges in detecting deepfakes is the high quality and realism of synthetic media, which renders conventional detection techniques less effective. Therefore, machine learning techniques capable of recognizing subtle patterns in visual data that are imperceptible to the human eye are required. This study aims to develop a deepfake image detection system using a hybrid machine learning approach that combines ResNet50 for feature extraction and XGBoost for classification. The pre-trained ResNet50 model, originally trained on the large-scale ImageNet dataset, is utilized to extract visual representations from images in the form of feature vectors. These features are then classified using XGBoost to distinguish between authentic and AI-generated images based on subtle patterns embedded within the extracted features. The results demonstrate that this hybrid approach achieves an accuracy of 94.6% in detecting deepfake images by leveraging the deep representation power of CNNs and the advanced classification capabilities of XGBoost. This method is not only computationally efficient but also highly relevant for integration into adaptive digital security systems.
Deepfake Audio Detection Using Hybrid Audio Spectrogram Transformer with Dual-Attention Feature Fusion and Bayesian Optimization Kusnaeni Kusnaeni; Muhaimin Hading; Zaitun Zaitun; Andi Oxy Raihan Machikami R; Afif Budi Andy B
Journal of Mathematics, Computations and Statistics Vol. 9 No. 3 (2026): Volume 09 Issue 03 (September 2026)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/ja96za26

Abstract

Deepfake audio has become a serious threat to digital security due to its ability to generate highly realistic synthetic speech for malicious purposes such as impersonation, fraud, and social engineering. Despite the rapid advancement of deepfake generation technologies, audio-based detection remains relatively underexplored compared to image and video domains. Moreover, existing approaches largely rely on handcrafted features and conventional machine learning methods, which are often insufficient to capture complex temporal and spectral dependencies and often exhibit limited generalization to unseen deepfake generation techniques. This study proposes a novel framework that leverages log-Mel spectrogram representations to capture informative time–frequency characteristics of audio signals. The proposed model integrates a Hybrid Audio Spectrogram Transformer to model global dependencies through self-attention and is further enhanced by a Dual-Attention Feature Fusion mechanism to improve discriminative capability at both spatial and channel levels. In addition, Bayesian Optimization is employed to efficiently optimize hyperparameters, leading to improved model performance, faster convergence, and better generalization. Experimental results demonstrate that the proposed model achieves an accuracy of 93.60% and an AUC of 0.9724, indicating excellent discriminative performance. The model also achieves a high recall in detecting deepfake audio, highlighting its effectiveness in minimizing false negatives and improving robustness in real-world scenarios. These findings demonstrate that the proposed framework provides an effective and reliable solution for deepfake audio detection and offers significant potential for deployment in practical cybersecurity and digital audio forensic applications.
Co-Authors Andi Oxy Raihan Machikami R Hading, Muhaimin Ika Reskiana Adriani Irwan Usman Kusnaeni Kusnaeni Mega Sartika Hafid Muh. Rifandi Muhammad Edy Rizal Muhammad Hidayatullah Zaitun Zaitun