Article Per Year (5 Year)
p-Index From 2020 - 2025
0.444
P-Index
This Author published in this journals
All Journal Krea-TIF: Jurnal Teknik Informatika Building of Informatics, Technology and Science Scientific Journal of Informatics
Hamim Wigena, Aji
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Engineering Library & Information Science

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

Found 3 Documents
Search

 1 
PERBANDINGAN METODE KEKAR BIWEIGHT MIDCOVARIANCE DAN MINIMUM COVARIANCE DETERMINANT DALAM ANALISIS KORELASI KANONIK Riana, Freza; Hamim Wigena, Aji; ., Erfiani
Krea-TIF: Jurnal Teknik Informatika Vol 3 No 2 (2015)
Publisher : Fakultas Teknik dan Sains, Universitas Ibn Khaldun Bogor

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (475.8 KB) | DOI: 10.32832/kreatif.v3i2.410

Abstract

Canonical Correlation Analysis(CCA) is a multivariate linear used toidentify and quantify associationsbetween two sets of random variables. Itsstandard computation is based on samplecovariance matrices, which are howeververy sensitive to outlying observations.The robust methods are needed. Thereare two robust methods, i.e robustBiweight Midcovariance (BICOV) andMinimum Covariance Determinant(MCD) methods. The objective of thisresearch is to compare the performanceof both methods based on mean squareerror. The data simulations aregenerated from various conditions. Thevariation data consists of the proportionof outliers, and the kind of outliers: shift,scale, and radial outlier. Theperformance of robust BICOV method inCCA is the best compared to MCD andClassic
Classification Modeling with RNN-based, Random Forest, and XGBoost for Imbalanced Data: A Case of Early Crash Detection in ASEAN-5 Stock Markets Siswara, Deri; M. Soleh, Agus; Hamim Wigena, Aji
Scientific Journal of Informatics Vol. 11 No. 3: August 2024
Publisher : Universitas Negeri Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15294/sji.v11i3.4067

Abstract

Purpose: This research aims to evaluate the performance of several Recurrent Neural Network (RNN) architectures, including Simple RNN, Gated Recurrent Units (GRU), and Long Short-Term Memory (LSTM), compared to classic algorithms such as Random Forest and XGBoost, in building classification models for early crash detection in the ASEAN-5 stock markets. Methods: The study examines imbalanced data, which is expected due to the rarity of market crashes. It analyzes daily data from 2010 to 2023 across the major stock markets of the ASEAN-5 countries: Indonesia, Malaysia, Singapore, Thailand, and the Philippines. A market crash is the target variable when the primary stock price indices fall below the Value at Risk (VaR) thresholds of 5%, 2.5%, and 1%. Predictors include technical indicators from major local and global markets and commodity markets. The study incorporates 213 predictors with their respective lags (5, 10, 15, 22, 50, 200) and uses a time step of 7, expanding the total number of predictors to 1,491. The challenge of data imbalance is addressed with SMOTE-ENN. Model performance is evaluated using the false alarm rate, hit rate, balanced accuracy, and the precision-recall curve (PRC) score. Result: The results indicate that all RNN-based architectures outperform Random Forest and XGBoost. Among the various RNN architectures, Simple RNN is the most superior, primarily due to its simple data characteristics and focus on short-term information. Novelty: This study enhances and extends the range of phenomena observed in previous studies by incorporating variables such as different geographical zones and periods and methodological adjustments.
Evaluasi Performa Rmixmod dan KAMILA dalam Pengelompokan Perguruan Tinggi di Indonesia Berdasarkan Data Capaian Kinerja Bertipe Campuran Santoso, Andrianto; Kurnia, Anang; Hamim Wigena, Aji
Building of Informatics, Technology and Science (BITS) Vol 7 No 1 (2025): June (2025)
Publisher : Forum Kerjasama Pendidikan Tinggi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47065/bits.v7i1.7376

Abstract

Clustering is a technique for grouping objects based on their similarities within clusters and their differences across clusters. In real-world, objects often have characteristics represented by a combination of numerical and categorical variables, requiring clustering techniques that can process mixed-type data. Model-based clustering is one of the approaches that can be utilized for such data. This study evaluates and compares two model-based clustering algorithms for mixed data type, Rmixmod, which employs a mixture model with maximum likelihood estimation and expectation-maximization, and KAMILA, which utilizes a semi-parametric approach. Both algorithms are implemented to cluster Indonesian higher education institutions based on their performance. The optimal number of clusters is determined using the Bayesian Information Criterion and the Silhouette Coefficient. Algorithms performance is evaluated using the Silhouette Coeeficient, the Calinski-Harabasz Index, and the Davies-Bouldin Index. The research results showed that the Rmixmod algorithm outperformed KAMILA in clustering Indonesian higher education institutions, with a Silhouette Coeeficient of 0.2878, a Calinski-Harabasz Index of 253.9433, and a Davies-Bouldin Index of 1.5321. The optimal number of clusters formed was five. Cluster interpretation is conducted by analyzing the mean values of PC and the distribution of categorical variables within each cluster. The clustering results are expected to serve as a foundation for the government in formulating strategic policies that are both effective and differentiated according to the characteristics of each group of higher education institutions.
Co-Authors Agus Mohamad Soleh Anang Kurnia Deri Siswara Erfiani Erfiani Freza Riana Santoso, Andrianto