Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Jurnal Teknologi dan Manajemen Informatika Jurnal Ilmiah Betrik : Besemah Teknologi Informasi dan Komputer
Miftahul Falah
Universitas Sriwijaya
Computer Science & IT Decision Sciences, Operations Research & Management Engineering

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

Found 2 Documents
Search

 1 
Model Matematika Hibrida Lexicon–Harris Hawks Optimization untuk Analisis Sentimen Ulasan Produk Shopee Miftahul Falah; Yesinta Florensia; Dewi Sartika
BETRIK Vol. 16 No. 03 (2025): Jurnal Ilmiah BETRIK : Besemah Teknologi Informasi dan Komputer
Publisher : PPPM Institut Teknologi Pagar Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36050/h239eh92

Abstract

This research presents a hybrid Lexicon–Harris Hawks Optimization (HHO) mathematical model designed to improve sentiment analysis performance on Shopee product reviews. The rapid growth of e-commerce platforms has resulted in a large volume of user-generated content, making accurate sentiment classification essential for understanding customer opinions. Traditional lexicon-based methods are simple and interpretable but often limited by static sentiment scores that fail to capture contextual nuances in real-world review data. To overcome these limitations, this study integrates a lexicon-based scoring approach with HHO to dynamically optimize sentiment weights and enhance classification accuracy. The proposed method involves four main stages: data preprocessing, baseline lexicon-based sentiment scoring, lexicon weight optimization using HHO, and final sentiment classification. HHO is employed to search for optimal lexicon weight configurations through exploration and exploitation mechanisms modeled after the cooperative hunting behavior of Harris hawks. The optimized weights are then applied to recalculate sentiment scores and classify reviews into positive, negative, or neutral categories. Performance is evaluated using accuracy, precision, recall, and F1-score. Experimental results show that the hybrid model significantly outperforms the baseline lexicon method, achieving near-perfect classification performance. The confusion matrix reveals extremely low misclassification rates, while evaluation metrics exceed 90% across all categories. The convergence curve further demonstrates stable and efficient optimization behavior. Analysis of sentiment score outputs indicates improved sensitivity for both positive and negative expressions, as well as more accurate representation of reviewer intent.
Comparative Analysis of SMOTE, WMOTE, and ADASYN Oversampling Methods on Multinomial Naive Bayes Performance in Classifying Toddlers Nutritional Status Naretha Kawadha Pasemah Gumay; Dewi Sartika; Rendra Gustriansyah; Yesinta Florensia; Miftahul Falah
Jurnal Teknologi dan Manajemen Informatika Vol. 12 No. 1 (2026): Juni 2026
Publisher : Universitas Merdeka Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Class imbalance in toddler nutritional status data often reduces the ability of classification models, especially in predicting minority classes. This study aims to analyze the impact of three oversampling techniques, namely SMOTE, WMOTE, and ADASYN, on improving the performance of the Multinomial Naive Bayes (MNB) algorithm. A dataset of 243 data was processed through a preprocessing stage by converting categorical variables using numeric labels. To meet the MNB algorithm's requirement for non-negative data, continuous numeric features (such as birth weight, birth height, weight, height, and age) were normalized using the Min-Max Scaler to the range [0, 1]. This process discretizes continuous values onto a probability scale to ensure feature compatibility with the Multinomial distribution. Data balancing was performed only on the training dataset, where the SMOTE method produced 374 data, ADASYN produced 375 data, and WMOTE produced 373 data. The evaluation results show that although all three oversampling methods experienced a slight decrease in global accuracy, the model's ability to detect minority classes improved, as evidenced by increases in G-Means and Balanced Accuracy. The test results concluded that MNB-ADASYN was the best model for prioritizing high sensitivity to all class labels, while MNB-WMOTE provided the most consistent global accuracy stability while maintaining performance on minority classes.
Co-Authors Dewi Sartika Naretha Kawadha Pasemah Gumay Rendra Gustriansyah Yesinta Florensia