Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal JOURNAL OF APPLIED INFORMATICS AND COMPUTING Jasmine : Journal of Intelligent Systems and Machine Learning
Very Setiawan
Universitas Pignatelli Triputra
Computer Science & IT

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Deep Learning Model for Identification of Indonesian National Figure Entities on Social Media Using LSTM Architecture Very Setiawan; Dwi Utari Iswavigra; Mutia Ulfa
JASMINE: Journal of Intelligent Systems and Machine Learning Vol. 1 No. 1 (2026)
Publisher : Universitas Telkom

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25124/jasmine.v1i1.10057

Abstract

In the era of rapid digital communication, social media has become a dominant medium for information exchange and public discourse, particularly in Indonesia. Despite this growth, automatic identification of national figures within social media texts remains a significant challenge due to the informal nature of language, frequent abbreviations, and inconsistent spelling patterns. Addressing this gap, this study aims to develop a Deep Learning model based on Long Short-Term Memory (LSTM) networks to identify Indonesian national figures from social media texts. The research utilizes 1,109 tweets collected from X (formerly Twitter) through the X API, encompassing names of well-known figures from politics, sports, entertainment, and social activism. The research process includes dataset crawling, preprocessing, labeling using the spaCy library, dividing training and test data, and training an LSTM model. The evaluation results show that the proposed model achieves a high level of performance, achieving 97.8% accuracy, 96% precision, 93% recall, and an F1-score of 92% on the validation data, demonstrating the LSTM model's ability to make accurate and reliable predictions. Word cloud analysis shows that the model is able to consistently recognize person entities such as "Prabowo", "Sri Mulyani", and "Agnez Mo". However, the model still experiences limitations in detecting unfamiliar or rarely appearing entities. Overall, this study shows that the combination of spaCy and LSTM is effective for NER tasks on Indonesian social media texts and has the potential for further development with increased data variety and improvements to the labeling process.
Performance Analysis of BERT and CLIP Models in Multimodal Sentiment Classification of Short Video Content Very Setiawan; Endang Anggiratih; Najwa Eka Putriningsih; Jonathan Eldo Kusuma
Journal of Applied Informatics and Computing Vol. 10 No. 3 (2026): June 2026
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v10i3.12822

Abstract

The rapid growth of short video platforms such as YouTube Shorts has increased the need for effective sentiment analysis methods capable of capturing public opinion in multimodal content. This study analyzes and compares the effectiveness of unimodal and multimodal approaches for sentiment classification of Indonesian short videos, focusing on IndoBERT for text-based modeling and CLIP for multimodal integration. The main objective is to investigate whether incorporating visual information alongside textual data can improve sentiment classification performance compared to a text-only approach. The dataset consists of 1,128 Indonesian short videos collected from YouTube Shorts. Audio data are transcribed into text using Automatic Speech Recognition (ASR), while visual information is represented using video thumbnails. Sentiment labels are automatically categorized into three classes (positive, neutral, and negative) using a pre-trained IndoBERT model. In the training phase, the unimodal approach relies solely on textual features extracted by IndoBERT, whereas the multimodal approach integrates textual and visual features using CLIP through feature-level fusion. Model performance is evaluated using accuracy, precision, recall, F1-score, and computational time analysis. The experimental results show that the unimodal text-based model outperforms the multimodal model, achieving higher accuracy (86% vs 82%) and better overall evaluation metrics. IndoBERT also demonstrates better convergence behavior compared to English BERT, with training accuracy increasing from 0.76 to 0.86 and validation accuracy from 0.77 to 0.88, along with lower loss values. In contrast, English BERT achieves lower performance, with training accuracy rising from 0.72 to 0.79 and validation accuracy from 0.73 to 0.80. Furthermore, the unimodal approach requires significantly less computation time (18 minutes compared to 35 minutes). These findings indicate that textual information plays a dominant role in sentiment expression in Indonesian short video content, while visual features increase computational complexity without significant performance gains.
Co-Authors Dwi Utari Iswavigra Endang Anggiratih Jonathan Eldo Kusuma Mutia Ulfa Najwa Eka Putriningsih