Claim Missing Document
Check
Articles

Found 10 Documents
Search

Klasifikasi Kategori Cuaca Berdasarkan Citra Menggunakan VGG-16 Riyadi, Sugeng; Pardede, Doughlas; Fuad, Raja Nasrul
Data Sciences Indonesia (DSI) Vol. 4 No. 1 (2024): Article Research Volume 4 Issue 1, June 2024
Publisher : ITScience (Information Technology and Science)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/dsi.v4i1.4664

Abstract

Penelitian ini bertujuan untuk mengklasifikasikan kondisi cuaca seperti Berawan, Cerah, dan Terbit, menggunakan citra digital dengan pendekatan otomatis berbasis Convolutional Neural Network (CNN). Arsitektur VGG-16 dipilih karena kemampuannya dalam mengekstraksi fitur detail melalui lapisan konvolusional bertingkat. Dataset yang digunakan berisi 910 citra, dibagi menjadi tiga kategori, dan diolah menggunakan VGG-16 untuk menghasilkan vektor fitur berdimensi 4096. Klasifikasi dilakukan dengan jaringan saraf tiruan yang memiliki tiga lapisan tersembunyi, dan evaluasi model menggunakan metode 10-fold cross-validation. Metrik yang digunakan untuk menilai kinerja model adalah akurasi, presisi, dan recall. Hasil penelitian menunjukkan bahwa VGG-16 mampu mengklasifikasikan citra dengan akurasi sebesar 96,48%, dengan performa terbaik pada kelas Berawan, Cerah, dan Terbit, yaitu masing-masing 96%, 95,5%, dan 97,2%. Meskipun model menunjukkan akurasi tinggi, tantangan masih ada dalam membedakan citra dengan fitur visual yang serupa, seperti intensitas cahaya dan formasi awan. Kesimpulannya, VGG-16 efektif dalam klasifikasi kondisi cuaca berbasis citra digital, namun memerlukan pengembangan lebih lanjut untuk mengatasi kesalahan klasifikasi akibat kemiripan visual antara kategori cuaca
Perbandingan Kinerja Kernel SVM dalam Klasifikasi Kategori Kanker Kulit Menggunakan Transfer Learning Siregar, Muhammad Mizan; Hizria, Rahmatika; Pardede, Doughlas
Data Sciences Indonesia (DSI) Vol. 4 No. 1 (2024): Article Research Volume 4 Issue 1, June 2024
Publisher : ITScience (Information Technology and Science)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/dsi.v4i1.4665

Abstract

Penelitian ini mengkaji efektivitas kombinasi transfer learning dengan Support Vector Machines (SVM) untuk klasifikasi kanker kulit. Tujuan utama dari penelitian ini adalah untuk mengklasifikasikan gambar kulit secara akurat ke dalam kelas "jinak" dan "ganas". Model VGG-19 yang telah dilatih sebelumnya digunakan untuk mengekstraksi fitur-fitur mendalam dari gambar kulit, menangkap pola visual yang rumit. Fitur-fitur ini kemudian dimasukkan ke dalam classifier SVM, dengan eksplorasi dilakukan pada kernel Radial Basis Function (RBF) dan Polynomial. Kinerja model-model yang diusulkan dievaluasi menggunakan dataset gambar kulit. Hasil eksperimen menunjukkan bahwa SVM dengan kernel Polynomial mengungguli SVM dengan kernel RBF dalam hal akurasi dan recall, khususnya untuk kelas "ganas". Hal ini menunjukkan bahwa kernel Polynomial lebih baik dalam menangkap hubungan kompleks dalam data. Pendekatan transfer learning, yang memanfaatkan model VGG-19 yang telah dilatih sebelumnya, secara signifikan meningkatkan kemampuan model untuk mengekstraksi fitur yang bermakna dari gambar, berkontribusi pada peningkatan akurasi klasifikasi. Temuan ini menunjukkan bahwa kombinasi transfer learning dan SVM, terutama dengan kernel Polynomial, menawarkan pendekatan yang menjanjikan untuk klasifikasi kanker kulit. Metode yang diusulkan dapat membantu dalam deteksi dini penyakit dan meningkatkan akurasi diagnosis, yang berpotensi mengarah pada hasil pasien yang lebih baik. Penelitian di masa depan dapat mengeksplorasi penggunaan dataset yang lebih besar dan lebih beragam, serta integrasi fitur atau teknik tambahan untuk lebih meningkatkan kinerja klasifikasi.
Klasifikasi Kategori Cuaca Berdasarkan Citra Menggunakan VGG-16 Riyadi, Sugeng; Pardede, Doughlas; Fuad, Raja Nasrul
Data Sciences Indonesia (DSI) Vol. 4 No. 1 (2024): Article Research Volume 4 Issue 1, June 2024
Publisher : Yayasan Cita Cendikiawan Al Kharizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/dsi.v4i1.4664

Abstract

Penelitian ini bertujuan untuk mengklasifikasikan kondisi cuaca seperti Berawan, Cerah, dan Terbit, menggunakan citra digital dengan pendekatan otomatis berbasis Convolutional Neural Network (CNN). Arsitektur VGG-16 dipilih karena kemampuannya dalam mengekstraksi fitur detail melalui lapisan konvolusional bertingkat. Dataset yang digunakan berisi 910 citra, dibagi menjadi tiga kategori, dan diolah menggunakan VGG-16 untuk menghasilkan vektor fitur berdimensi 4096. Klasifikasi dilakukan dengan jaringan saraf tiruan yang memiliki tiga lapisan tersembunyi, dan evaluasi model menggunakan metode 10-fold cross-validation. Metrik yang digunakan untuk menilai kinerja model adalah akurasi, presisi, dan recall. Hasil penelitian menunjukkan bahwa VGG-16 mampu mengklasifikasikan citra dengan akurasi sebesar 96,48%, dengan performa terbaik pada kelas Berawan, Cerah, dan Terbit, yaitu masing-masing 96%, 95,5%, dan 97,2%. Meskipun model menunjukkan akurasi tinggi, tantangan masih ada dalam membedakan citra dengan fitur visual yang serupa, seperti intensitas cahaya dan formasi awan. Kesimpulannya, VGG-16 efektif dalam klasifikasi kondisi cuaca berbasis citra digital, namun memerlukan pengembangan lebih lanjut untuk mengatasi kesalahan klasifikasi akibat kemiripan visual antara kategori cuaca
Perbandingan Kinerja Kernel SVM dalam Klasifikasi Kategori Kanker Kulit Menggunakan Transfer Learning Siregar, Muhammad Mizan; Hizria, Rahmatika; Pardede, Doughlas
Data Sciences Indonesia (DSI) Vol. 4 No. 1 (2024): Article Research Volume 4 Issue 1, June 2024
Publisher : Yayasan Cita Cendikiawan Al Kharizmi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/dsi.v4i1.4665

Abstract

Penelitian ini mengkaji efektivitas kombinasi transfer learning dengan Support Vector Machines (SVM) untuk klasifikasi kanker kulit. Tujuan utama dari penelitian ini adalah untuk mengklasifikasikan gambar kulit secara akurat ke dalam kelas "jinak" dan "ganas". Model VGG-19 yang telah dilatih sebelumnya digunakan untuk mengekstraksi fitur-fitur mendalam dari gambar kulit, menangkap pola visual yang rumit. Fitur-fitur ini kemudian dimasukkan ke dalam classifier SVM, dengan eksplorasi dilakukan pada kernel Radial Basis Function (RBF) dan Polynomial. Kinerja model-model yang diusulkan dievaluasi menggunakan dataset gambar kulit. Hasil eksperimen menunjukkan bahwa SVM dengan kernel Polynomial mengungguli SVM dengan kernel RBF dalam hal akurasi dan recall, khususnya untuk kelas "ganas". Hal ini menunjukkan bahwa kernel Polynomial lebih baik dalam menangkap hubungan kompleks dalam data. Pendekatan transfer learning, yang memanfaatkan model VGG-19 yang telah dilatih sebelumnya, secara signifikan meningkatkan kemampuan model untuk mengekstraksi fitur yang bermakna dari gambar, berkontribusi pada peningkatan akurasi klasifikasi. Temuan ini menunjukkan bahwa kombinasi transfer learning dan SVM, terutama dengan kernel Polynomial, menawarkan pendekatan yang menjanjikan untuk klasifikasi kanker kulit. Metode yang diusulkan dapat membantu dalam deteksi dini penyakit dan meningkatkan akurasi diagnosis, yang berpotensi mengarah pada hasil pasien yang lebih baik. Penelitian di masa depan dapat mengeksplorasi penggunaan dataset yang lebih besar dan lebih beragam, serta integrasi fitur atau teknik tambahan untuk lebih meningkatkan kinerja klasifikasi.
Kajian Literatur Multi Layer Perceptron Seberapa Baik Performa Algoritma Ini Pardede, Doughlas; Hayadi, B.Herawan; Iskandar
Journal of ICT Applications System Vol 1 No 1 (2022): Journal of ICT Aplications and System
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (306.323 KB) | DOI: 10.56313/jictas.v1i1.127

Abstract

Multi Layer Perceptron (MLP), one of the deep learning algorithms, has been widely used in classification problem research because it has advantages over other conventional classification methods. This study takes 15 articles that have been published in research journals regarding the application of the multi-layer perceptron algorithm to prediction and classification problems. From the results of the analysis carried out, the results show that the lowest performance value of this algorithm is 62.89%, the highest performance value of this algorithm is 100% and the average performance value of this algorithm is 91.98%. From these values, it can be concluded that the multi layer perceptron algorithm is very good and feasible to be used in solving prediction and classification problems
MotoGP Mandalika 2022 Sentiment Classification Using Machine Learning Pardede, Doughlas; Hayadi, B. Herawan
Jurnal Transformatika Vol. 20 No. 2 (2023): January 2023
Publisher : Jurusan Teknologi Informasi Universitas Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26623/transformatika.v20i2.5364

Abstract

MotoGP is a world-class motorcycle racing event, which will be held in the 19th series in 2022 at the Pertamina Mandalika Circuit. This study tries to analyze public sentiment collected from the results of tweeter social media tweets, in the form of sentiment and emotion values. With the features of sentiment and emotion values extracted from the contents of this tweet, k-means clustering is used to generate sentiment clusters as targets for classification using the MLP algorithm. From the results of the evaluation using 10-fold cross validation, the accuracy value is 97%, the precision value is 94.64% and the recall value is 100%. The classification results also show that the public response to the 2022 MotoGP event at the Mandalika circuit is quite balanced, where 53% have a positive response, while the rest have a negative response
ANALISIS DAMPAK STRATEGI PEDAGOGI TERHADAP MINAT BELAJAR SISWA MENGGUNAKAN RANDOM FOREST Sinaga, Novendra Adisaputra; Pardede, Doughlas; Riyadi, Sugeng
Jurnal Teknik Informasi dan Komputer (Tekinkom) Vol 8 No 1 (2025)
Publisher : Politeknik Bisnis Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37600/tekinkom.v8i1.2169

Abstract

Student learning interest plays a crucial role in educational success, as it directly influences engagement, comprehension, and academic achievement. This study aims to analyze the influence of pedagogical strategies on students’ learning interest using a machine learning approach with the Random Forest algorithm. Eight aspects of teaching strategies were examined as predictor variables, while learning interest was measured through two main indicators: interest in real-world application of the material and motivation for self-directed learning. Data were collected from 100 students via a Likert-scale questionnaire and analyzed using Orange Data Mining. The model was validated through 10-fold cross-validation and evaluated using accuracy, precision, recall, F1-score, and AUC. The results indicate strong model performance, with 95% accuracy, 96.7% precision, 97.8% recall, and a 97.2% F1-score. Feature importance analysis identified practical activities (P4), an inclusive learning environment (P6), and the use of technology (P3) as the most influential predictors of learning interest. In contrast, variables such as P1, P2, and P8 showed minimal contribution. These findings demonstrate that Random Forest is not only effective for classification tasks but also valuable in identifying key factors for improving pedagogical strategies. The results are expected to inform the development of more adaptive, interactive, and student-centered learning environments.
Enhancing Multi-Layer Perceptron Performance with K-Means Clustering Pardede, Doughlas; Ichsan, Aulia; Riyadi, Sugeng
Journal of Computer Networks, Architecture and High Performance Computing Vol. 6 No. 1 (2024): Article Research Volume 6 Issue 1, January 2024
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v6i1.3600

Abstract

Machine learning plays a crucial role in identifying patterns within data, with classification being a prominent application. This study investigates the use of Multilayer Perceptron (MLP) classification models and explores preprocessing techniques, particularly K-Means clustering, to enhance model performance. Overfitting, a common challenge in MLP models, is addressed through the application of K-Means clustering to streamline data preparation and improve classification accuracy. The study begins with an overview of overfitting in MLP models, highlighting the significance of mitigating this issue. Various techniques for addressing overfitting are reviewed, including regularization, dropout, early stopping, data augmentation, and ensemble methods. Additionally, the complementary role of K-Means clustering in enhancing model performance is emphasized. Preprocessing using K-Means clustering aims to reduce data complexity and prevent overfitting in MLP models. Three datasets - Iris, Wine, and Breast Cancer Wisconsin - are employed to evaluate the performance of K-Means as a preprocessing technique. Results from cross-validation demonstrate significant improvements in accuracy, precision, recall, and F1 scores when employing K-Means clustering compared to models without preprocessing. The findings highlight the efficacy of K-Means clustering in enhancing the discriminative power of MLP classification models by organizing data into clusters based on similarity. These results have practical implications, underlining the importance of appropriate preprocessing techniques in improving classification performance. Future research could explore additional preprocessing methods and their impact on classification accuracy across diverse datasets, advancing the field of machine learning and its applications
Analysis of Logistic Regression Regularization in Wild Elephant Classification with VGG-16 Feature Extraction Ichsan, Aulia; Riyadi, Sugeng; Pardede, Doughlas
Journal of Computer Networks, Architecture and High Performance Computing Vol. 6 No. 2 (2024): Articles Research Volume 6 Issue 2, April 2024
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v6i2.3789

Abstract

The research article explores the intersection of image-based wildlife classification and logistic regression regularization, focusing on the classification of wild elephant species. It begins by highlighting the significance of ecological research in biodiversity monitoring and conservation and introduces Convolutional Neural Networks (CNNs) as potent tools for feature extraction from images. The VGG-16 model is particularly emphasized for its ability to capture hierarchical representations of visual features crucial for classification tasks. The integration of VGG-16 feature extraction with logistic regression regularization is proposed as a compelling approach, offering a balance between sophisticated feature representation and efficient classification algorithms. The literature review delves into image-based wildlife classification, emphasizing the role of CNNs, especially VGG-16, in extracting discriminative features. It discusses the fusion of VGG-16 features with logistic regression and the challenges in this field, such as dataset annotation and environmental variability. The method section outlines the dataset acquisition, feature extraction using the VGG-16 architecture, and model configuration using logistic regression with lasso and ridge regularization. The process of finding the optimal regularization parameter (lambda) and model evaluation through cross-validation is detailed. Results showcase the optimal lambda values for lasso and ridge regularization and compare the performance of logistic lasso and logistic ridge models. Misclassification analysis reveals factors influencing classification accuracy, including feature variability and contextual complexity. The discussion reflects on the implications of the findings, emphasizing the importance of lambda selection and addressing challenges in wildlife classification. It suggests avenues for further research, such as advanced modeling techniques and feature engineering approaches. In conclusion, the study contributes to advancing wildlife classification efforts by leveraging state-of-the-art techniques and sheds light on opportunities to enhance classification accuracy in wildlife conservation.
Digital Green Education through Green Chemistry Supports the Green Economy by Improving Science Skills and Entrepreneurial Character Sitompul, Hamela Sari; Maulina, Intan; Pardede, Doughlas
Jurnal Penelitian Pendidikan IPA Vol 11 No 10 (2025): October
Publisher : Postgraduate, University of Mataram

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29303/jppipa.v11i10.12332

Abstract

Green education is integrated into chemistry learning to support the green economy, as it offers numerous applications in everyday life, including waste management and sustainable reforestation programs. Based on the Merdeka Curriculum, one aspect of chemistry learning in grade 10 is green chemistry, which explores global issues and problem-solving. Science skills are closely linked to green education in chemistry learning, as students gain scientific knowledge and attitudes from the theories they learn. All these efforts are directed toward achieving the Sustainable Development Goals (SDGs). The research used a quasi-experimental Design. The study had two groups, the experimental and the control groups. The sample was SMA Negeri 1 Gunung Meria, Deli Serdang Regency students. The research instruments used were essay tests and observations. The results of the posttest t-test analysis between the control and the experimental group at a significance level of 0.05, then 0.00 <0.05, then Ho is rejected and Ha is accepted. The results of the students' entrepreneurial character scores can be seen in Figure 1, where the experimental class has an average score of 77.70 and the control class 42.85. The results of the study can be concluded that education has a significant influence on students' science process skills. Students' entrepreneurial character shows a substantial difference, in that in the experimental class, there is good development.