Article Per Year (5 Year)
p-Index From 2021 - 2026
1.633
P-Index
This Author published in this journals
All Journal IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Seminar Nasional Aplikasi Teknologi Informasi (SNATI) Proceedings of KNASTIK TELKOMNIKA (Telecommunication Computing Electronics and Control) Jurnal Informatika Jurnal Sarjana Teknik Informatika Rekursif: Jurnal Informatika Komuniti : Jurnal Komunikasi dan Teknologi Informasi Jurnal Pengabdian UntukMu NegeRI Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) EDUMATIC: Jurnal Pendidikan Informatika Building of Informatics, Technology and Science Mobile and Forensics Aviation Electronics, Information Technology, Telecommunications, Electricals, Controls (AVITEC) Jurnal Teknik Informatika (JUTIF) ABDIMASTEK
Dewi Soyusiawaty
Universitas Ahmad Dahlan
Religion Aerospace Engineering Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Industrial & Manufacturing Engineering Law, Crime, Criminology & Criminal Justice Library & Information Science Mechanical Engineering Neuroscience Social Sciences

Published : 37 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 37 Documents
Search

 1   2   3   4 
SMOTE-SVM for Handling Imbalanced Data in Obesity Classification Biddinika, Muhammad Kunta; Yuliansyah, Herman; Soyusiawaty, Dewi; Razak, Farhan Radhiansyah
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 19, No 2 (2025): April
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.103994

Abstract

 Obesity is a significant health issue associated with various chronic diseases, making its early classification critical for effective interventions. This study investigates the performance of Support Vector Machine (SVM) models with Radial Basis Function (RBF) and Linear kernels on imbalanced obesity datasets. To address data imbalance, Synthetic Minority Over-sampling Technique (SMOTE) and Random Undersampling (RUS) were applied. The results reveal that balancing techniques significantly enhance classification performance, with the Linear model achieving the highest accuracy of 96.54% when balanced using SMOTE. However, limitations include reduced recall for minority classes and potential overfitting risks. These findings underscore the importance of balancing techniques in health data classification and offer insights for further optimizing model performance. The study highlights the need for advanced data balancing strategies to improve predictive accuracy and equity across all classes.
Klasifikasi Jenis Kejahatan berdasarkan Teks Amar Putusan Pengadilan Hukum Pidana KUHP menggunakan IndoBERT Perdana, Tirtanusa Kurnia Adhi; Soyusiawaty, Dewi
Jurnal Pendidikan Informatika (EDUMATIC) Vol 9 No 2 (2025): Edumatic: Jurnal Pendidikan Informatika
Publisher : Universitas Hamzanwadi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29408/edumatic.v9i2.30326

Abstract

The increasing number of the court’s rulings each year presents a challenge for the judiciary. One strategic solution is the application of Artificial Intelligence (AI). Indonesian-based models such as IndoBERT is potential to ease workloads by automatically classifying legal cases. This study aims to explore the capability of IndoBERT to automatically classifying the verdict of section of Indonesian KUHP rulings to accelerate crime type identification. This is an experimental study using supervised text classification. The dataset consists of 12000 verdicts collected from the Indonesian Supreme Court website, classified using IndoBERT fine-tuned with various hyperparameter configuration. Our findings show that the model with a batch size of 8 and learning rate 5e-5 achieved accuracy of 92.59%, precison of 92.93%, recall of 92.59%, and F1-Score of 92.59% on unseen test data. The high accuracy is supported by the explicit mention of crime types within verdict texts. To date, no study has specifically utilized IndoBERT or other models for automatic classification of KUHP articles. This finding has the potential to be integrated into the Supreme Court’s Directory of Decision as a support tool for automatic classification and legal document archiving.
Analisis Sentimen Program Makan Siang Gratis di Twitter/X menggunakan Metode BI-LSTM Attaulah, Dimas Thaqif; Soyusiawaty, Dewi
Jurnal Pendidikan Informatika (EDUMATIC) Vol 9 No 1 (2025): Edumatic: Jurnal Pendidikan Informatika
Publisher : Universitas Hamzanwadi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29408/edumatic.v9i1.29725

Abstract

The free lunch program became a widely discussed topic on social media, reflecting public opinion towards the policy. This research aims to analyze public sentiment towards free lunch program to evaluate the policy's effectiveness and understand public perception. Data was collected through web crawling techniques on the Twitter/X platform, resulting in 7,441 data. Processing stages include preprocessing, sentiment labeling using VADER, keyword visualization with wordcloud, and application of word embedding using Word2Vec. The oversampling technique is used to overcome data imbalance. Sentiment classification was developed using Bi-LSTM and evaluated with accuracy, precision, recall, and F1-score. The developed Bi-LSTM model achieved 88.75% accuracy, with 88.9% precision, 88.8% recall, and 88.8% F1-score. Analysis results show that the majority of public responses are positive or neutral, although there were negative sentiments that highlighted potential problems such as corruption and increasing national debt. These results provide insight into public opinion on the free lunch policy and demonstrate the effectiveness of the Bi-LSTM model in social media sentiment classification.
Pelatihan Kreasi Konten Digital dengan Komunikasi melalui Tools Kecerdasan Artifisial Winiarti, Sri; Soyusiawaty, Dewi; Umar, Rusydi; Yuliansyah, Herman
Jurnal Pengabdian UntukMu NegeRI Vol. 9 No. 3 (2025): Pengabdian Untuk Mu negeRI
Publisher : LPPM UMRI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37859/jpumri.v9i3.10417

Abstract

Seiring dengan adanya kebijakan Kementrian Pendidikan Dasar dan Menengah Republik Indonesia terkait penerapan Kecerdasan Artifisial (KA) dalam pembelajaran jenjang Sekolah Dasar (SD) hingga Sekolah Menengah Atas (SMA), maka semua sekolah memerlukan adanya pemahaman terhadap pelaksanaan pembelajaran KA. Perkembangan KA telah memberikan peluang baru dalam mendukung kreativitas dan komunikasi digital dalam pelaksanaan pembelajaran. Namun, banyak guru masih kesulitan berinteraksi secara efektif dengan perangkat KA untuk menghasilkan konten pembelajaran yang meraik, khususnya untuk pembelajaran dengan model unplugged. Kegiatan pengabdian kepada masyarakat ini bertujuan meningkatkan kompetensi guru dalam berkomunikasi dengan perangkat KA melalui pelatihan bertema “Kreasi Konten Digital dengan Komunikasi melalui Tools Kecerdasan Artifisial.” Pelatihan dilaksanakan pada 9 Juli 2025 di Kabupaten Sleman dengan peserta sebanyak 24 guru Sekolah Menengah Pertama (SMP). Metode yang digunakan meliputi tranfer pengetahuan, praktik langsung Aplikasi KA, praktek mengajar dengan Aplikasi KA dengan pendekatan PjBL dan evaluasi pelaksanaan. Materi pelatihan berupa konsep KA dalam pembelajaran, cara berkomunikasi dengan perangkat KA yang efektif dengan menggunakan prompt yang efektif. Kegiatan pelatihan ini meningkatkan keterampilan komunikasi guru sebesar 17,4% (dari 69,4% menjadi 86,8%), menunjukkan efektivitas pendekatan PjBL dalam memperkuat literasi digital dan kreativitas guru.
Spell Correction for the Minangkabau Language Using BERT-Based Embeddings Dewi Soyusiawaty; Abdul Fadlil; Sunardi
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 10 No 3 (2026): June 2026
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29207/resti.v10i3.7182

Abstract

Spell checking is an essential component of natural language processing, as it directly influences applications such as sentiment analysis, text classification, and machine translation. Developing a reliable system for low-resource languages like Minangkabau is challenging due to frequent spelling variations and limited annotated data. This study proposes a contextual spell correction model using pre-trained IndoBERT and multilingual BERT (mBERT) embeddings applied without additional training. The method masks misspelled words, extracts the contextual embedding of the [MASK] token, and compares it with candidate embeddings generated through dictionary filtering and Levenshtein Distance. Evaluation was conducted on the Spell Error Corpus for Minangkabau Language (SPEML), which includes insertion errors, deletion errors, substitution errors, transposition errors, punctuation errors, real-word errors, and loanword errors. Results show that mBERT consistently outperformed IndoBERT, achieving an average F1-score of 0.83 compared to 0.75. Statistical validation using paired t-test and Wilcoxon signed-rank test further confirmed that the performance difference between the two models was significant. Both models reached perfect scores (1.0) in real-word and loanword categories, and strong results in insertion_medium (0.97 for mBERT and 0.95 for IndoBERT). The lowest performance occurred in deletion_short (0.52 for IndoBERT) and long words cases (0.57 for mBERT). In addition, a small-scale external validation using 100 Twitter/X sentences was conducted to assess the applicability of the proposed method to real-world social media text. Overall, the findings confirm the effectiveness of contextual embeddings for Minangkabau spelling correction while highlighting challenges in long misspelled words, deletion errors, and informal real-world text.
Sentiment Analysis Of E-Commerce Reviews Using Fine-Tuned Indobert With Class Weights Strategy Syakura, Abdan; Soyusiawaty, Dewi
Jurnal Teknik Informatika (Jutif) Vol. 7 No. 3 (2026): JUTIF Volume 7, Number 3, June 2026
Publisher : Informatika, Universitas Jenderal Soedirman

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52436/1.jutif.2026.7.3.5654

Abstract

MSMEs in the e-commerce sector face difficulties in converting large volumes of unstructured customer review data into actionable business insights. This challenge is exacerbated by the ambiguity of star ratings, which often do not align with the content of the reviews, making automated sentiment analysis of the text essential. This study implements a systematic sentiment analysis workflow on a case study of 15,278 customer reviews of Toko Pasar Stan Jogja. The method used is fine-tuning a pre-trained Transformer model, namely IndoBERT, which is optimized with class weighting techniques to handle unbalanced datasets. The model's performance was comprehensively evaluated using Accuracy, Precision, Recall, F1-Score, Confusion Matrix, and word cloud visualization metrics. The test results showed that the developed model had very high performance, achieving an overall accuracy of 96.99% and an average F1-Score of 0.97 on the test data. Qualitative analysis also successfully identified that product quality (“fresh”) and logistics efficiency (‘fast’) were the main drivers of satisfaction, while the main complaints centered on the condition of the product upon arrival (“damaged,” “rotten”). This research proves that the optimized Transformer model is not only effective for sentiment classification, but also serves as a strategic tool for extracting concrete business insights.
Pelatihan Aplikasi Keuangan Menggunakan Excel bagi Guru-Guru TK ABA Ngabean Lisna Zahrotun; Dewi Soyusiawaty; Ninda Khoirunnisa; Hana Jelita Sari
ABDIMASTEK Vol. 4 No. 2 (2025): Desember
Publisher : Universitas Muhammadiyah Jember

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

Abstract

Dalam mengelola keuangan organisasi maupun Lembaga lebih memanfaatkan sistem informasi keuangan, hal ini dilakukan karena mempermudah dalam pembukuan dan pelaporan keuangan. Banyak Lembaga pemerintahan yang sudah menggunakan aplikasi dalam pengelolaan keuangan salah satunya adalah Lembaga Pendidikan. Namun tidak semua jenjang Pendidikan sudah memiliki dan mampu menggunakan aplikasi keuangan. Salah satunya adalah TK ABA Ngabean yang masih menggunakan pencatatan manual daram pengelolaan keuangan sekolah. Tujuan dari pelatihan ini adalah mempermudah kegiatan pencatatan transaksi keluar masuk dan pelaporan keuangan. Aplikasi keuangan dibangun Excel Visual Basic for Application dan Macro. Hal ini disesuaikan dengan kemampuan pemahaman dari Guru dan Tenaga Administrasi TK ABA Ngabean. Selain itu aplikasi ini hanya membutuhakn perangkat yang sederhana melalui Microsoft excel. Pelatihan dilakukan selama 1 hari secara luring dan dilanjutkan pendampingan secara daring. Dari hasil pengujian aplikasi menggunakan metode SUS dengan 14 responden diperoleh nilai 71.6, artinya aplikasi ini layak dan dapat membantu pengelolaan keuangan terutama  pencataan transaksi dan pelaporan sekolah.
Co-Authors Abdan Syakura Abdul Fadlil Aldini Fujiarti Aning Setiya Rini Ardian M.N. Attaulah, Dimas Thaqif Buhari Muslim Catur Prasongko Devi Ayunani Wandasari Endri Rahmawanto Enggar Tyasto Wingwirawan Esty Istiqamah Failasuf Abid Dainuri Fathurrahman Fathurrahman Freddy Alpino Furqon, Ihtiar Nur Hana Jelita Sari Hapsari, Intan Herman Yuliansyah, Herman Irfan Nur Aditya Jones, Anna Hendri Soleliza Kholid Rohman Saputro Linda Mariana Lisna Zahrotun Muhammad Kunta Biddinika Ninda Khoirunnisa Nizar Anggit Pramana Nuri Piyanto Perdana, Tirtanusa Kurnia Adhi Ramadhan, Imam Razak, Farhan Radhiansyah Rosary Indah Dewi Rusydi Umar Sri Winiarti Sunardi, Sunardi