cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Muhammad Hijrah, M.Si
Contact Email
journaltimuris@gmail.com
Phone
+6282193229395
Journal Mail Official
journaltimuris@gmail.com
Editorial Address
Jl. J. Syarnamual-Ambon, Maluku-Indonesia
Location
Kota ambon,
Maluku
INDONESIA
Timuris : Journal of Computational and Information Research
Published by Kiswantara Academic Publishing
ISSN : -     EISSN : 31641571     DOI : -
Core Subject :
Timuris : Journal of Computational and Information Research is an open-access scholarly journal dedicated to advancing innovative research in the fields of computing, information technology, and intelligent systems. The journal publishes two issues annually, released in June and December. Established in 2026, TIMURIS serves as a platform for the dissemination of high-quality, original research articles that contribute significantly to the advancement of computer science and related disciplines. The journal covers a broad range of topics, including artificial intelligence, machine learning, data science, information systems, and applied computational technologies. All published articles are assigned a unique Digital Object Identifier (DOI) to ensure persistent accessibility and ease of citation. TIMURIS is registered with e-ISSN 3164-1571, and is managed using the Open Journal Systems (OJS) platform, providing seamless online access for researchers, academics, and practitioners worldwide. TIMURIS aims to foster scientific innovation, interdisciplinary collaboration, and knowledge dissemination, with a particular emphasis on supporting technological and research development in emerging and underrepresented regions, including Eastern Indonesia. For more detailed information regarding the journal’s focus and scope, authors and readers are encouraged to visit the Focus and Scope section on the journal’s website.
Arjuna Subject : -
Articles 5 Documents
 1 
Sentiment Analysis of TikTok Comments on the Weakening of the Rupiah Exchange Rate as an Indicator of Public Perception of Financial Risk Baiq Wira Hartati; Ahda Sabila Wulandari; Risma Anggraeni; Nur Asmita Purnamasari
Timuris: Journal of Computational and Information Research Vol. 1 No. 1 (2026): Timuris: Journal of Computational and Information Research
Publisher : Kiswah Institute

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

Abstract

The weakening of the rupiah against the U.S. dollar is an economic issue that can influence the public’s perception of financial risk. The social media platform TikTok has become one of the channels the public uses to express their opinions on current economic conditions. This study aims to analyze the sentiment of TikTok users’ comments regarding the weakening of the rupiah as an indicator of the public’s perception of financial risk. The method used is text mining with a Lexicon-Based Sentiment Analysis approach. The research data consists of 163 TikTok comments discussing the weakening of the rupiah exchange rate. The analysis stages include data preprocessing, word cloud visualization, and sentiment classification into positive, neutral, and negative categories. The results show that negative sentiment dominates at 56.44%, followed by neutral sentiment at 38.65%, and positive sentiment at 4.91%. The most frequently occurring words include “Prabowo,” “MBG,” “rupiah,” “president,” “rise,” and “dollar.” The dominance of negative sentiment indicates public concern regarding the impact of the weakening rupiah, such as rising prices, declining purchasing power, and economic uncertainty. The research results suggest that social media sentiment analysis can serve as an early indicator for understanding the public’s perception of financial risks in real time.
Comparison of Historical Simulation and Variance-Covariance Methods for Value at Risk Estimation of BBRI Stock Jihan Afifah; Nanda Aulia Sudiasmini; Nur Aminingsih; Nur Asmita Purnamasari
Timuris: Journal of Computational and Information Research Vol. 1 No. 1 (2026): Timuris: Journal of Computational and Information Research
Publisher : Kiswah Institute

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

Abstract

Stock investment is exposed to market risk arising from fluctuations in stock prices. Therefore, accurate risk measurement is essential for investors and risk managers. Among the various tools available for quantifying investment risk, Value at Risk (VaR) has gained widespread adoption as a method for determining the worst expected loss under a given probability threshold. This study compares the Historical Simulation and Variance-Covariance methods in estimating the Value at Risk of PT Bank Rakyat Indonesia (Persero) Tbk (BBRI) stock using daily closing price data from January 2, 2024, to December 31, 2025. The Jarque-Bera normality test indicated that the return data were not normally distributed, suggesting the presence of non-normal characteristics in the return distribution. Based on an assumed investment value of IDR 10,000,000, the VaR estimates at the 95% confidence level were IDR 334,739 and IDR 356,444 using Historical Simulation and Variance-Covariance, respectively. At the 99% confidence level, the estimated VaR values were IDR 534,695 and IDR 500,158, respectively. Kupiec Proportion of Failures (POF) backtesting showed that both methods produced statistically valid VaR estimates. However, Historical Simulation generated a more conservative risk estimate at the 99% confidence level, indicating a greater ability to capture extreme losses under non-normal return distributions. Therefore, Historical Simulation is recommended as the preferred method for measuring the market risk of BBRI stock.
Development of a Stability-Dispersion Adaptive Weighted K-Means Method for Feature-Sensitive Clustering Ramli Rumeon; Surman Siloyanan
Timuris: Journal of Computational and Information Research Vol. 1 No. 1 (2026): Timuris: Journal of Computational and Information Research
Publisher : Kiswah Institute

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

Abstract

This study develops Stability-Dispersion Adaptive Weighted K-Means (SDAW-K-Means), an extension of classical K-Means that updates feature weights according to within-cluster dispersion. Classical K-Means treats all standardized features equally, although some features may be more relevant for cluster separation than others. The proposed method estimates feature weights iteratively: features with smaller within-cluster dispersion receive larger weights, while less informative features receive smaller weights. The empirical illustration uses the public Iris dataset from the UCI Machine Learning Repository through scikit-learn. Results show that the proposed weighting mechanism is interpretable and can improve agreement with reference labels based on the adjusted Rand index. The article contributes a transparent feature-weighted K-Means formulation for applied clustering research.
Application of Principal Component Analysis (PCA) to Identify the Main Factors Causing Stunting Novica Sintasyah Sinaga; Elisabeth Princess; Aisyah Novianti
Timuris: Journal of Computational and Information Research Vol. 1 No. 1 (2026): Timuris: Journal of Computational and Information Research
Publisher : Kiswah Institute

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

Abstract

Stunting is a serious public health problem in Indonesia, including in North Sumatra Province where the prevalence is still above the threshold set by WHO. This study aims to identify the main factors that cause stunting in North Sumatra Province using the Principal Component Analysis (PCA) method. PCA is applied to reduce a number of variables that cause stunting into several main components that are able to explain the diversity of data to the maximum and overcome the problem of multicollinearity between variables. The data used is secondary data from stunting vulnerability indicators in districts/cities throughout North Sumatra Province. The results of the analysis showed that PCA succeeded in reducing the data dimension and identifying the main components with the highest eigenvalues that were the dominant factors causing stunting. These findings are expected to provide a comprehensive overview of the factors that have the most influence on stunting incidence in North Sumatra, so that it can be the basis for more targeted and effective intervention policy recommendations for local governments in an effort to accelerate the reduction of stunting rates.  
Classifying Family Economic Status Using the K-Nearest Neighbor Algorithm in Popalia Village Istrikah Istrikah; Rabiah Adawiyah; Yuwanda Purnamasari Pasrun
Timuris: Journal of Computational and Information Research Vol. 1 No. 1 (2026): Timuris: Journal of Computational and Information Research
Publisher : Kiswah Institute

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

Abstract

Access to accurate family economic data is essential for the equitable distribution of village social assistance. At the Popalia Village Office, Tanggetada Sub-district, Kolaka Regency, identification of eligible recipients previously relied on manual, page-by-page verification of Statistics Indonesia (BPS) census documents, a process that was slow and often produced recipients that did not match the intended criteria. This study develops a web-based classification system using the K-Nearest Neighbor (KNN) algorithm to categorize 160 household heads into “Mampu” (financially capable) and “Tidak Mampu” (financially incapable) classes based on twelve socio-economic criteria, including occupation, monthly income, education, number of dependents, and asset ownership. The system was built following the Waterfall development model using PHP and MySQL with a use-case-driven UML design. Model performance was evaluated using Euclidean-distance-based KNN with 10-fold cross validation and confusion matrix analysis. The system achieved an average classification accuracy of 99.38% (minimum 93.75%, maximum 100%), a precision of 98.21%, a recall of 100%, and an F1-score of 99.10%. Black-box testing further confirmed that all functional modules operated as intended. These findings indicate that KNN is an accurate and practical method for supporting village-level social assistance targeting.

Page 1 of 1 | Total Record : 5

 1 

Filter by Year

2026 2026


Reset
Filter By Issues
All Issue Vol. 1 No. 1 (2026): Timuris: Journal of Computational and Information Research