Garuda - Garba Rujukan Digital

Comparison of Supervised Learning Classification Methods on Accreditation Data of Private Higher Education Institutions Noviyanto; Wahyudi, Mochamad; Sumanto, Sumanto
Paradigma - Jurnal Komputer dan Informatika Vol. 26 No. 1 (2024): March 2024 Period
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/p.v26i1.3306

This research aims to analyze and compare supervised learning classification methods using a case study of accreditation data for private higher education institutions within the LLDikti Region III contained in BAN-PT. In addition, this research also uses Weka machine learning software in its calculations. The initial step taken is to prepare the software used for supervised learning analysis, then pre-processing the data, namely labeling data that has a categorical data type, after that determining data for testing data. The next step is to test each classification method. The methods used for comparison are logistic regression, K-nearest neighbor, naive bayes, super vector machine, and random forest. Based on the calculation results, the Kappa Statistic and Root mean squared error values obtained are 1 and 0 for the logistic regression method, 0.979 and 0.0061 for the K-nearest neighbor method, 1 and 0.2222 for the super vector machine method, 0.969 and 0.0341 for the naive bayes method, 1 and 0 for the decision tree method, and 0.5776 and 0.1949 for the random forest method, respectively. The logistic regression and decision tree methods in this study get Kappa Statistic and Root mean squared error values of 1 and 0 respectively so that they are said to be good and acceptable, thus the two classification methods are the most appropriate methods and are considered to have the highest accuracy.

Biodiversity and Analysis of Antioxidant and Antibacterial Activity of Endophytic Fungi Extracts Isolated from Mangrove Avicennia marina Noviyanto; Widjajanti, Hary; Elfita
Science and Technology Indonesia Vol. 10 No. 1 (2025): January
Publisher : Research Center of Inorganic Materials and Coordination Complexes, FMIPA Universitas Sriwijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26554/sti.2025.10.1.139-151

Avicennia marina, a mangrove species commonly found along coastal areas, plays both ecological and pharmacological roles, with its plant parts exhibiting antioxidant and antibacterial activities. This study aimed to investigate the diversity of endophytic fungi from various organs of A. marina collected from mangrove ecosystems and to explore and analyze their antioxidant and antibacterial activities. Endophytic fungi were isolated from the roots, stems, and fruits of A. marina using PDA medium and were morphologically identified. Each fungal isolate was cultivated in PDB medium for 4 weeks under static conditions, followed by extraction to obtain concentrated extracts. Antioxidant and antibacterial activities were assessed using the DPPH method and disk diffusion assay. A total of 23 fungal isolates were obtained from the roots, stems, and fruits of A. marina. The identification results showed that the root isolates had the highest genus diversity, followed by the stem and fruit isolates. The highest distribution of antioxidant and antibacterial activities was observed in the endophytic fungal extracts from fruits, followed by those from roots and stems. Notably, the majority of the 23 endophytic fungal extracts exhibited strong antioxidant and antibacterial activities. Isolates AMF3 and AMF6showed the most potent antioxidant activity, classified as very strong, with IC50 values below 20 ug/mL. Morphological identification revealed AMF3 as Neopestalotiopsis sp. and AMF6 as Aspergillus niger. This study highlights the potential of Neopestalotiopsis sp. and Aspergillus niger endophytic fungi from A. marina fruits as sources of natural antioxidant and antibacterial compounds, offering valuable insights for biotechnological applications of mangrove-associated endophytes.

Comparison of Supervised Learning Classification Methods on Accreditation Data of Private Higher Education Institutions Noviyanto; Wahyudi, Mochamad; Sumanto, Sumanto
Paradigma - Jurnal Komputer dan Informatika Vol. 26 No. 1 (2024): March 2024 Period
Publisher : LPPM Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/p.v26i1.3306

This research aims to analyze and compare supervised learning classification methods using a case study of accreditation data for private higher education institutions within the LLDikti Region III contained in BAN-PT. In addition, this research also uses Weka machine learning software in its calculations. The initial step taken is to prepare the software used for supervised learning analysis, then pre-processing the data, namely labeling data that has a categorical data type, after that determining data for testing data. The next step is to test each classification method. The methods used for comparison are logistic regression, K-nearest neighbor, naive bayes, super vector machine, and random forest. Based on the calculation results, the Kappa Statistic and Root mean squared error values obtained are 1 and 0 for the logistic regression method, 0.979 and 0.0061 for the K-nearest neighbor method, 1 and 0.2222 for the super vector machine method, 0.969 and 0.0341 for the naive bayes method, 1 and 0 for the decision tree method, and 0.5776 and 0.1949 for the random forest method, respectively. The logistic regression and decision tree methods in this study get Kappa Statistic and Root mean squared error values of 1 and 0 respectively so that they are said to be good and acceptable, thus the two classification methods are the most appropriate methods and are considered to have the highest accuracy.

BIOACTIVITY OF ENDOPHYTIC FUNGI EXTRACT ISOLATED FROM THE LEAVES OF MISTLETOE (Dendrophthoe pentandra (L.) Miq.) ON THE LIME PLANT (Citrus aurantifolia) Hiras Habisukan, Ummi; Oktiansyah, Rian; Noviyanto; Anjeli, Riri
Berita Biologi Vol 23 No 3 (2024): Berita Biologi
Publisher : BRIN Publishing (Penerbit BRIN)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55981/berita_biologi.2024.7423

Mistletoe (Dendrophthoe pentandra (L.) Miq.) is a parasitic plant used in traditional medicine. This study evaluates the bioactivity of endophytic fungi in mistletoe . Endophytic fungal species were determined through morphological identification. Potato Dextrose Broth (PDB) media was used for the cultivation, and ethyl acetate was used as solvent to extract secondary metabolites. The antioxidant test was carried out using the DPPH method, while the paper disc diffusion method performed the antibacterial test. A total of 4 isolates of endophytic fungi were obtained from mistletoe leaves, namely isolates DB1 – DB4. The results of morphological analysis showed that DB1 was Paecilomyces sp., DB2 was Papulaspora sp., DB3 was Aspergillus sp., and DB4 was Mucor sp. The endophytic fungus DB3 (Aspergillus sp.) showed the most potential antioxidant and antibacterial activity. This endophytic fungal extract can potentially be a source of new drugs through further research by isolating the active compound.

Comparison of CNN Transfer Learning Models for Brain Tumor Detection Based on MRI Images noviyanto; Pamuja, Sintia Darma
JKTI Jurnal Keilmuan Teknologi Informasi Vol 1 No 2 (2025)
Publisher : Universitas Muhammadiyah Klaten

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61902/jkti.v1i2.2185

Brain tumors require early and accurate detection to support effective clinical decision-making. This study compares the performance of four transfer learning-based Convolutional Neural Network (CNN) models, namely DenseNet121, InceptionV3, MobileNet, and Xception, for brain tumor detection using MRI images. The dataset was preprocessed through resizing, normalization, and data augmentation, and all models were trained for 20 epochs using ImageNet pre-trained weights. Model performance was evaluated using accuracy, precision, recall, and F1-score metrics. The experimental results show that all models achieved accuracies above 90%, with MobileNet outperforming the others by achieving an accuracy of 94.74% and precision, recall, and F1-score values of 0.95, 0.95 and 0,94. These findings indicate that lightweight CNN architectures can deliver superior performance for MRI-based brain tumor classification.

Optimasi Analisis Sentimen Ulasan Platform Pendidikan Daring Menggunakan Arsitektur ALBERT dan Teknik Augmentasi Kontekstual Pamuja, Sintia Darma; Noviyanto
JKTI Jurnal Keilmuan Teknologi Informasi Vol 1 No 2 (2025)
Publisher : Universitas Muhammadiyah Klaten

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61902/jkti.v1i2.2256

Online learning through global platforms like Coursera generates a massive volume of user reviews, which serve as vital information for educational quality improvement. However, these reviews often exhibit imbalanced label distributions, where positive sentiments significantly dominate negative and neutral ones, hindering traditional classification models. Advanced language models such as ALBERT (A Lite BERT) offer parameter efficiency through cross-layer parameter sharing while maintaining high performance in complex text understanding. This study aims to evaluate the ALBERT model's performance in classifying Coursera user reviews and addressing data imbalance using Contextual Word Embedding augmentation. The methodology involves collecting 10,000 reviews followed by preprocessing steps including case folding, punctuation removal, and tokenization. The augmentation technique utilizes language models to replace words based on context to balance minority classes. The results show that ALBERT provides highly consistent performance, achieving an F1-score of 0.9710 with the contextual augmentation scenario. The model proves effective in capturing linguistic variations and remains computationally efficient. In conclusion, the ALBERT model is highly effective for sentiment analysis on the Coursera dataset, where contextual augmentation significantly enhances the model's ability to recognize minority classes that were previously difficult to identify.

Title

Found 6 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 6 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 6 Documents
Search