Article Per Year (5 Year)
p-Index From 2020 - 2025
0.444
P-Index
This Author published in this journals
All Journal IAES International Journal of Artificial Intelligence (IJ-AI) JOIV : International Journal on Informatics Visualization
Suhali, Jason
Unknown Affiliation
Computer Science & IT Engineering

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

Found 2 Documents
Search

 1 
Dealing imbalance dataset problem in sentiment analysis of recession in Indonesia Kristiyanti, Dinar Ajeng; Sanjaya, Samuel Ady; Tjokro, Vinsencius Christio; Suhali, Jason
IAES International Journal of Artificial Intelligence (IJ-AI) Vol 13, No 2: June 2024
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijai.v13.i2.pp2060-2072

Abstract

Global recession news dominates social media, particularly in Indonesia, with social news platforms on Twitter generating public responses and re-tweetings on the issue. Mining these opinions from Twitter using a sentiment analysis approach yields invaluable insights. The research stages included data collection, pre-processing, data labeling using the lexical-based method like valence aware dictionary and sentiment reasoner (VADER) and TextBlob, sampling techniques using synthetic minority oversampling technique (SMOTE) and random over sampling (ROS) before and after splitting data, and modeling using machine learning such as support vector machines (SVM), k-nearest neighbour (KNN), naive Bayes, and model evaluation. The problem is that almost 300,000 data collected from NodeXL are unbalanced. The findings show that models with balanced datasets show better model evaluation results. The sampling technique was carried out before and after splitting the data. The model evaluation results show that the Bernoulli-naive Bayes algorithm, with the VADER labeling technique, and the SMOTE sampling technique after splitting data, obtains the best accuracy of 84%, and using the ROS technique obtains an accuracy of 81%. On the other hand, with the SMOTE and ROS technique before splitting data on the SVM algorithm, it gets the best accuracy of 93% from before if only using SVM only reached 84%.
Comparison of Salp Swarm Algorithm and Particle Swarm Optimization as Feature Selection Techniques for Recession Sentiment Analysis in Indonesia Kristiyanti, Dinar Ajeng; Sanjaya, Samuel Ady; Irmawati, Irmawati; Ekachandra, Kristian; Suhali, Jason; Hairul Umam, Akhmad
JOIV : International Journal on Informatics Visualization Vol 9, No 5 (2025)
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/joiv.9.5.3102

Abstract

Amidst global economic uncertainty, this study focuses on Twitter sentiment during the global recession issue on social media, especially in Indonesia. By utilizing sentiment analysis, this study uses machine learning algorithms such as Naïve Bayes (NB), Support Vector Machine (SVM), K-Nearest Neighbor (KNN) which are still less than optimal on high-dimensional Twitter data. The purpose of this study is to improve the accuracy of conventional machine learning using basic metaheuristic algorithms, namely the Salp Swarm Algorithm (SSA) and Particle Swarm Optimization (PSO) as feature selection. From January to May 2023, this study captures the evolving sentiment in response to economic conditions. Data preprocessing, including labeling through the TextBlob and VADER libraries, sets the stage for the analysis. Performance is compared based on labeling techniques, feature selection, and classification algorithms. Specifically, when applied to VADER labeled data without feature selection, the SVM model achieves an outstanding accuracy of 83% and an F1 score of 67%—notably, the application of SSA and PSO results in a reduction in model accuracy by 1%. However, the application of SSA and PSO slightly reduced the model accuracy performance by 1%. On the TextBlob labeled data, SVM showed an outstanding performance (80% accuracy, 77% F1 score). Interestingly, PSO on TextBlob data with SVM significantly decreased the model's performance. These findings contribute significantly to understanding the intricacies of sentiment dynamics during economic uncertainty on social media platforms, with SVM emerging as a strong choice for practical sentiment analysis.
Co-Authors Dinar Ajeng Kristiyanti, Dinar Ajeng Ekachandra, Kristian Hairul Umam, Akhmad Irmawati Irmawati Sanjaya, Samuel Ady Tjokro, Vinsencius Christio