cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Is Fatimah
Contact Email
eksakta@uii.ac.id
Phone
+6282326298724
Journal Mail Official
eksakta@uii.ac.id
Editorial Address
Faculty of Mathematics and Natural Sciences Universitas Islam Indonesia Jl. Kaliurang Km 14, Ngaglik, Sleman, Yogyakarta, 55584
Location
Kab. sleman,
Daerah istimewa yogyakarta
INDONESIA
EKSAKTA: Journal of Sciences and Data Analysis
Published by Universitas Islam Indonesia
ISSN : 27160459     EISSN : 27209326     DOI : 10.20885
Core Subject : Science, Engineering,
Chemical Engineering, Chemistry & Bioengineering Chemistry Earth & Planetary Sciences Materials Science & Nanotechnology
Ekstakta is an interdisciplinary journal with the scope of mathematics and natural sciences that is published by Fakultas MIPA Universitas Islam Indonesia. All submitted papers should describe original, innovatory research, and modelling research indicating their basic idea for potential applications. The Journal particularly welcomes submissions that focus on the progress in the field of mathematics, statistics, chemistry, physics, biology and pharmaceutical sciences.
Arjuna Subject : Kimia(semua kategori) - Kimia (Lain-Lain)
Articles 11 Documents
 1   2 
Search results for , issue "VOLUME 7, ISSUE 1, April 2026" : 11 Documents clear
Orange Classification using Naïve Bayes and K-Nearest Neighbor Algorithms based on Its Physical Properties Hafizulhaq, Fadli; Andasuryani, Andasuryani
EKSAKTA: Journal of Sciences and Data Analysis VOLUME 7, ISSUE 1, April 2026
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20885/EKSAKTA.vol7.iss1.art5

Abstract

Oranges are among the most widely consumed fruits globally. While many farmers possess extensive knowledge of orange cultivation, they often lack expertise in post-harvest handling and processing. Classification or grading is a crucial step after harvest to ensure quality. Machine learning offers an efficient solution for automating this process and decreasing the time consumed. This study implements two machine learning algorithms, Naïve Bayes and K-Nearest Neighbor, to classify Gerga oranges based on different training-to-test data ratios (75:25, 50:50, and 25:75). The results indicate that as the training data decreases, the accuracy of Naïve Bayes improves, but its precision declines, whereas K-Nearest Neighbor exhibits the opposite trend. The best accuracy (90% accuracy) was produced by NB-25 and KNN-75. Meanwhile, precision and recall value were more important in order to reduce economic losses and buyer dissatisfaction, so that users can profit more. In this case, the KNN-75 model is the best to classify Gerga oranges into theright groups (85% precision, 91% recall). Despite the differences in class importance, KNN offers a steadier and more balanced outcome for both sides of the dataset. KNN is also more reliable to handle many number of samples in real practice when the model is used to design sorting or grading machines for oranges.
Multi-state Models for Longitudinal Data with Hidden Markov Method Amritha, Yadhurani Dewi; Danarnodo, Danarnodo
EKSAKTA: Journal of Sciences and Data Analysis VOLUME 7, ISSUE 1, April 2026
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20885/EKSAKTA.vol7.iss1.art1

Abstract

A person’s health condition can certainly change from time to the time. Changes in this condition can be formed into a model, one of model is a multi-state with Markov assumptions. The live expectancy value of a person suffering from a chronic disease is never 100 % correct because there is a lot of uncertainty in the future. However, by selecting the right method, the expected value can be determined with a low error rate or provide the best possible estimate of the future state. A multi-state Hidden Markov Model (HMM) is utilized in this study to analyze longitudinal data on Type 2 Diabetes Mellitus, chosen specifically for its robust capacity to manage data collected with regular, irregular, or continuous observation schedules. This model is also used to estimate the transition and observation probabilities with the maximum likelihood method. Additionally, estimates for the transition intensity and transition probability were calculated for each of the four possible model specifications. From the models that can be formed, the best model is determined through the AIC value. In this case, the best model is the model that uses covariates in each transition
Analisis Regresi Binomial Negatif terhadap Faktor-Faktor yang Memengaruhi Kasus Tuberkulosis di Jawa Barat, Indonesia Fikriya, Aufa; Shalfa Salsabilla; Raisah Zharifah Labibah; Sri Winarni; Defi Yusti Faidah; Anindya Apriliyanti Pravitasari; Triyani Hendrawati; Irlandia Ginanjar
EKSAKTA: Journal of Sciences and Data Analysis VOLUME 7, ISSUE 1, April 2026
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20885/EKSAKTA.vol7.iss1.art3

Abstract

Tuberculosis (TB) remains a major public health problem globally, with West Java reporting the highest number of TB cases among all provinces in Indonesia in 2023. This study aims to identify key factors influencing TB incidence across districts and cities in West Java in 2024. The analysis focuses on healthy living behaviors, proper sanitation, HIV cases, and AIDS cases using a Negative Binomial Regression approach to address overdispersion in count data. The results show that proper sanitation has a significant negative association with TB incidence, while HIV and AIDS cases exhibit significant positive associations. The best-performing model includes these three variables, yielding a residual deviance of 27.615. These findings highlight the importance of integrated public health interventions that simultaneously improve sanitation and strengthen HIV/AIDS control programs to effectively reduce TB incidence in high-burden regions.
Acute Toxicity Evaluation of Phyllanthus niruri SNEDDS Formulation According to OECD 425 Hikmah, Uzulul; Ramadani, Arba Pramundita; Canastie, Putri Julia; Ma'wa, Nurul Jannatul
EKSAKTA: Journal of Sciences and Data Analysis VOLUME 7, ISSUE 1, April 2026
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20885/EKSAKTA.vol7.iss1.art2

Abstract

A self-nanoemulsifying drug delivery system (SNEDDS) is a preparation that can increase the solubility of poorly soluble compounds, including Phyllanthus niruri. However, increasing solubility and bioavailability carries the risk of increased toxicity. Previous research showed that SNEDDS of Phyllanthus niruri (SNPN) at a dose of 100 mg/kgBW has hepatoprotective activity through decreasing ALT and AST enzyme levels after administering paracetamol at a dose of 3 g/kgBW in rats. However, at a dose of 200 mg/kg, ALT and AST levels increased, indicating potential toxicity. This study aims to determine the LD50 and to conduct histopathological observations of the liver and kidneys following oral administration of SNPN, in accordance with OECD 425 guidelines. Dose determination follows the recommendations of the AOT 425 StatPgm software, which includes the limit test (2000 mg/kg BW) and the main test (175, 550, and 2000 mg/kg BW), with observations made over 14 days. Next, the animals were sacrificed and necropsied to collect the liver and kidney organs for histopathological observation. LD50 determination was performed using AOT 425 StatPgm, and the level of organ damage was assessed using histopathological scoring. The result showed that the LD₅₀ value of SNPN exceeded 2000 mg/kgBW, placing it in category 5 (mild toxicity). Histopathological tests showed varying severity of changes between doses, indicating a toxic effect of SNPNadministration on the microscopic structure of the target organs. In conclusion, SNPN administration has the potential to cause mild acute toxicity, as indicated by the LD50 results and histological changes in the liver and kidneys.
Pengelompokan Sektor Pipa Minyak dan Gas dengan Metode K-Medoid Berbasis Blok Dimas Zahran Wicaksana; Kariyam, Kariyam; Suryanto, Tri
EKSAKTA: Journal of Sciences and Data Analysis VOLUME 7, ISSUE 1, April 2026
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20885/EKSAKTA.vol7.iss1.art7

Abstract

Effective oil and gas pipeline management requires a data-driven approach to identify segments with varying risk characteristics. This study aims to classify pipeline segments based on protective infrastructure conditions using the Block-Based K-Medoids clustering method. The analysis considers six variables: Pipeline Burial, Pipe Along Road, Pipe Guards, Berm/Rail/Guard Condition, Public Road, and ROW HCA along a 59-kilometer pipeline corridor. Data were normalized, and the optimal number of clusters was determined using the Deviation Ratio Index based on Medoid (DRIM), which indicated three clusters as the most representative structure. The results demonstrate clear differentiation among segments in terms of exposure level, protective condition, and HCA involvement, enabling classification into low-, moderate-, and high-risk groups. Spatial visualization further confirms systematic risk distribution along the route. These findings provide a structured basis for prioritizing inspection, maintenance, and mitigation strategies in pipeline infrastructure management.
Optimizing Medan Tourist Routes Using BiogeographyBased Optimization Zai, Fidelis Nofertinus; Nainggolan, Donni Andreas; Kurnia, Rian; -, Erwin
EKSAKTA: Journal of Sciences and Data Analysis VOLUME 7, ISSUE 1, April 2026
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20885/EKSAKTA.vol7.iss1.art8

Abstract

This study optimizes tourist routes across 14 destinations in the city of Medan using the Biogeography-Based Optimization (BBO) algorithm. The problem is formulated as a closed-path Traveling Salesman Problem (TSP) with an extension allowing for flexibility in freely selecting the starting point. The route is determined based on the distance between two locations, where the distance is assumed to be asymmetric to account for real-world urban road conditions such as one-way systems, while ignoring traffic conditions and other costs. Simulation results show that even though the starting point is freely determined, the BBO algorithm is still able to consistently produce routes that are close to optimal with stable convergence. The main contribution of this study is the provision of an adaptive and realistic route planning model to support tourism information systems in urban areas.
Treatment of Ginger Industrial Waste Using the Carbon Dioxide Supercritical Extraction (CDSE) Method Riyanto, Riyanto; Naeli Khusnul Maria; Elfara Rahmadhika Suwarno
EKSAKTA: Journal of Sciences and Data Analysis VOLUME 7, ISSUE 1, April 2026
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20885/EKSAKTA.vol7.iss1.art4

Abstract

Ginger industrial waste is available in abundance, unprocessed and harmful to the environment because of its smells. Researchers propose that ginger industrial waste be processed using water distillation and Carbon Dioxide Supercritical Extraction (CDSE) methods. The purpose of this study was to utilize ginger industrial waste, which is currently discarded. Ginger industrial waste was collected from the ginger beverage industry. Essential oil was extracted from the waste using two methods: water distillation and CDSE. Both essential oils were characterized to determine their chemical components and composition, as well as their physical and chemical properties in accordance with the Indonesian National Standard for ginger essential oil (SNI 1312:2021). The characteristics analyzed included specific gravity, refractive index, optical rotation, and chemical compound profile, as determinated using GC-MS. The yields of essential oil obtained from the water distillation and CDSE methods were 0.18% and 0.36%, respectively. The results of the specific gravity, refractive index, and optical rotation tests for the two methods did not differ significantly. The chromatogram profiles of the two methods also did not differ significantly. The study concluded that the CDSE method was superior to the water distillation method. The CDSE method is faster, requires no fuel, produces higher yields, and is environmentally friendly.
Inflation Convergence Modeling Using Binary Logistic Regression With SGD-Newton Raphson Optimization Methods in Indonesia Fatma Novalia Kussumarani; Istiqomah, Nerissabila Uswatun; Siva Ifin Azzahra; Anggraini Puspita Sari; Sischa Wahyuning Tyas
EKSAKTA: Journal of Sciences and Data Analysis VOLUME 7, ISSUE 1, April 2026
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20885/EKSAKTA.vol7.iss1.art9

Abstract

Global economic changes have necessitated the development of inflation models that can accurately describe Indonesia's economic dynamics. This study aims to compare two optimization methods, Newton Raphson and Stochastic Gradient Descent (SGD), in binary logistic regression modeling to analyze the effectiveness of monetary policy. This study contributes to evaluating the performance of both methods in terms of convergence speed and accuracy of inflation model parameter estimation. The results of the analysis show that the Newton Raphson method is more efficient in achieving convergence with an iteration value of 0.2933 compared to SGD, while both methods produce equivalent model quality based on the Akaike Information Criterion (AIC) values of 34.4008 and 34.4254. These findings emphasize the importance of selecting the right optimization method to support more efficient monetary policy analysis.
Long-Memory Modeling of Farmers' Terms of Trade in Indonesia: A Comparative Analysis of SARIMA and SARFIMA Approaches Viranty, Miftah Rizky; Rahkmawati, Yeni; Asianingrum, Al Hujjah
EKSAKTA: Journal of Sciences and Data Analysis VOLUME 7, ISSUE 1, April 2026
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20885/EKSAKTA.vol7.iss1.art6

Abstract

Indonesia, as an agrarian country, places the agricultural sector as a vital pillar of its economy and food security, with farmers’ welfare measured through the Farmers’ Terms of Trade (FTT). This study aims to compare the performance of the Seasonal Autoregressive Integrated Moving Average (SARIMA) and the Seasonal Autoregressive Fractionally Integrated Moving Average (SARFIMA) models in forecasting FTT using monthly data from 2009 to 2024 obtained from BPS (Statistics Indonesia). The results show that the SARIMA(0,1,1)(0,1,1)₁₂ model demonstrates higher accuracy with a MAPE value of 5.29%, compared to SARFIMA(1,0.2688,0)(0,1,1)₁₂ with a MAPE of 5.97%. However, the relatively small difference in MAPE indicates the presence of long-memory characteristics in the FTT data, although it does not significantly improve forecasting accuracy. The forecast results based on the best SARIMA model predict that FTT will gradually increase throughout 2025, peaking at 127.2920 in December, with a temporary decline from March to May. These findings can serve as a basis for the government to formulate targeted agricultural policies, price control measures, subsidy distribution, and marketing strategies that enhance farmers’ welfare and support national food security.
A Comparative Evaluation of XGBoost and LightGBM for Diabetes Mellitus Risk Prediction Using a Public Dataset and Web-Based Dashboard Wahyuningtyas, Sischa; Al Hafidz, Muhammad
EKSAKTA: Journal of Sciences and Data Analysis VOLUME 7, ISSUE 1, April 2026
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20885/EKSAKTA.vol7.iss1.art11

Abstract

Diabetes mellitus is a health problem of global concern, considering that most cases are only identified when complications arise. Therefore, early detection is essential in controlling the health and financial consequences of the disease. The purpose of this study is to compare two machine learning models using gradient boosting techniques, namely Extreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM). This study use a technique called RandomizedSearchCV to optimize the performance of the proposed machine learning models. In evaluating the machine learning models, the study used a variety of metrics such as accuracy, precision, recall, F1 score, and ROC-AUC. The LightGBM is a more efficient machine learning model than XGBoost based on the result. The LightGBM model had a classification accuracy of 77.3%, a precision of 71.1%, and a recall of 59.3%, which is the same value obtained by the XGBoost model. However, the LightGBM model had a higher F1 score of 64.6% and a ROC-AUC of 83.0% which indicates that the model is more balanced and can accurately classify and distinguish between the two classes. The best-performing machine learning model was integrated with a web-based system using a framework called Streamlit to create a system that is responsive, interactive, and user-friendly. The system is useful for early detection of diabetes mellitus and can be used by non-experts to determine whether a patient is at risk of developing the disease using real-time prediction and user-friendly data input. The results of the study showed that gradient boosting machine learning models can be used to diagnose and detect early cases of diabetes mellitus.

Page 1 of 2 | Total Record : 11

 1   2 

Filter by Year

2026 2026


Reset
Filter By Issues
All Issue VOLUME 7, ISSUE 1, April 2026 VOLUME 6, ISSUE 2, October 2025 VOLUME 6, ISSUE 1, April 2025 VOLUME 5, ISSUE 2, October 2024 VOLUME 5, ISSUE 1, April 2024 VOLUME 4, ISSUE 2, August 2023 VOLUME 4, ISSUE 1, February 2023 VOLUME 3, ISSUE 2, August 2022 VOLUME 3, ISSUE 1, February 2022 VOLUME 2, ISSUE 2, August 2021 VOLUME 2, ISSUE 1, February 2021 VOLUME 1, ISSUE 2, August 2020 VOLUME 1, ISSUE 1, February 2020 VOLUME 19, ISSUE 2, August 2019 VOLUME 19, ISSUE 1, February 2019 VOLUME 18, ISSUE 2, August 2018 VOLUME 18, ISSUE 1, February 2018 VOLUME 17, ISSUE 2, August 2017 VOLUME 17, ISSUE 1, February 2017 VOLUME 16, ISSUE 2, August 2016 VOLUME 16, ISSUE 1, February 2016 VOLUME 15, ISSUE 1-2, 2015 VOLUME 14, ISSUE 2, August 2014 VOLUME 14, ISSUE 1, February 2014 VOLUME 13, ISSUE 1-2, August 2013 VOLUME 13, ISSUE 1, February 2012 VOLUME 12, ISSUE 2, August 2011 VOLUME 12, ISSUE 1, February 2011 VOLUME 11, ISSUE 2, August 2010 VOLUME 11, ISSUE 1, February 2010 More Issue