Article Per Year (5 Year)
p-Index From 2020 - 2025
0.408
P-Index
This Author published in this journals
All Journal Communication in Biomathematical Sciences BACA: Jurnal Dokumentasi dan Informasi
Khotimah, Purnomo Husnul
Unknown Affiliation
Library & Information Science Social Sciences

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

Found 2 Documents
Search

 1 
Optimizing Multi-Layer Perceptron Performance in Sentiment Classification through Neural Network Feature Extraction Alam, Muhammad Fikri; Nuryaman, Aang; Khotimah, Purnomo Husnul; Parlina, Anne; Sihombing, Andre
BACA: Jurnal Dokumentasi dan Informasi Vol. 46 No. 1 (2025): BACA: Jurnal Dokumentasi dan Informasi (Juni)
Publisher : Direktorat Repositori, Multimedia, dan Penerbitan Ilmiah - Badan Riset dan Inovasi Nasional (BRIN Publishing)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55981/baca.2025.8240

Abstract

There are some problems with using the Multi-Layer Perceptron (MLP) model for complex tasks because it can be hard to understand hierarchical relationships and tends to overfit data with a lot of dimensions. This research proposes an enhanced MLP model for sentiment classification by integrating feature extraction layers from advanced neural networks, specifically the Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (Bi-LSTM). These layers aim to improve the model's representation capabilities by capturing more nuanced features. To evaluate the performance improvements of this augmented MLP model, metrics such as accuracy, precision, recall, F1-score, and the Area Under the Curve for Receiver Operating Characteristics (ROC-AUC) were employed. A key metric focus is the delta value, representing changes in the ROC-AUC, to assess the significance of these enhancements. The integration of CNN as a feature extraction layer yielded optimal ROC-AUC results, achieving values of 93.30% and 93.00%, which reflect an improvement of 0.51% and 4.46% over the baseline model. These findings indicate that adding feature extraction layers significantly enhances MLP performance in sentiment classification tasks. Future research may explore the potential of using alternative neural networks as feature extractors to continue advancing MLP capabilities in complex NLP applications.
Regresi Multiskala Tertimbang Geografis dan Temporal dengan LASSO dan Adaptif LASSO untuk Pemetaan Kejadian Tuberkulosis di Jawa Barat Habsy, Muhammad Yusuf Al; Rachmawati, Ro'fah Nur; Khotimah, Purnomo Husnul; Natari, Rifani Bhakti; Riswantini, Dianadewi; Munandar, Devi; Izzaturrahim, Muh. Hafizh
Communication in Biomathematical Sciences Vol. 8 No. 1 (2025)
Publisher : The Indonesian Bio-Mathematical Society

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.5614/cbms.2025.8.1.6

Abstract

Tuberculosis (TB) is a global health issue caused by Mycobacterium tuberculosis and can affect any organ of the body, especially the lungs. The trend of TB cases varies between regions, and analytic assessment is required to identify the predictor variables. The purpose of this research is to compare the Multiscale Geographically and Temporally Weighted Regression (MGTWR) and the Geographically and Temporally Weighted Regression (GTWR) method, which both use Gaussian, Exponential, Uniform, and Bi-Square kernel functions, to identify significant variables in each region annually. The MGTWR method has the advantage of using a flexible bandwidth for each observation, that results in more accurate coefficient estimates. The sample used was 27 districts and cities in West Java Province, involving 36 variables divided into 5 dimensions, namely global climate, health, demography, population, and government policy, with a time span of 2019–2022. To overcome the problem of multicollinearity, the approach was carried out using the Least Absolute Shrinkage Selection Operator (LASSO) and Adaptive LASSO methods. In determining the best model, the prioritized criteria are to achieve the highest R2, which indicates the optimal level of model fit, as well as the smallest AIC, which indicates the most efficient model goodness of fit. The best model is MGTWR with LASSO variable selection on the Bi-Square kernel. This model has an R2 of 91.25% and the smallest AIC of 139.868. From the best model, each region emerged with a cluster structure affected by various variables from 2019 to 2022, providing an in-depth understanding of TB mapping that can assist in formulating more effective intervention measures.
Co-Authors Aang Nuryaman Alam, Muhammad Fikri Anne Parlina Devi Munandar, Devi Habsy, Muhammad Yusuf Al Izzaturrahim, Muh. Hafizh Natari, Rifani Bhakti Rachmawati, Ro'fah Nur Riswantini, Dianadewi Sihombing, Andre