Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal Jurnal ELTIKOM : Jurnal Teknik Elektro, Teknologi Informasi dan Komputer JOURNAL OF APPLIED INFORMATICS AND COMPUTING
Ignasius, Darnell
Unknown Affiliation
Aerospace Engineering Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Engineering

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Gaussian Mixture-Based Data Augmentation Improves QSAR Prediction of Corrosion Inhibition Efficiency Ignasius, Darnell; Akrom, Muhamad; Budi, Setyo
Journal of Applied Informatics and Computing Vol. 9 No. 5 (2025): October 2025
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v9i5.10895

Abstract

Predicting corrosion inhibition efficiency IE (%) is often hindered by small, heterogeneous datasets. This study proposes a Gaussian mixture–based data augmentation pipeline to strengthen QSAR generalization under data scarcity. A curated set of 70 drug-like compounds with 14 physicochemical and quantum descriptors was cleaned, split 90/10 (train/test), and transformed using a Quantile Transformer followed by a Robust Scaler. A Gaussian Mixture model (GMM) with 2–5 components selected by the variational lower bound was fitted to the transformed training features and used to generate up to 2,500 synthetic samples. Eight regressors (Gaussian Process, Decision Tree, Random Forest, Bagging, Gradient Boosting, Extra Trees, SVR, and KNN) were evaluated on the held-out test set using R2 and RMSE. Augmentation improved performance across several families: for example, Gaussian Process R2 improved from −1.54 to 0.54 (RMSE 11.71 to 5.01) and Decision Tree R2 from −0.33 to 0.63 (RMSE 8.48 to 4.44), Bagging and Random Forest showed R2 increases of 0.67 and 0.40, respectively. The optimal synthetic size varied by model.
Medical Named Entity Recognition from Indonesian Health-News using BiLSTM-CRF with Static and Contextual Embeddings Ignasius, Darnell; Novita Dewi , Ika; Bernadette Chayeenee Norman , Maria; Rakhmat Sani, Ramadhan
Journal of Applied Informatics and Computing Vol. 9 No. 6 (2025): December 2025
Publisher : Politeknik Negeri Batam

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30871/jaic.v9i6.11574

Abstract

Named Entity Recognition (NER) is vital for structuring medical texts by identifying entities such as diseases, symptoms, and drugs. However, research on Indonesian medical NER remain limited due to the lack of annotated corpora and linguistic resources. This scarcity often leads to difficulties in learning meaningful word representations, which are crucial for accurate entity identification. This research aims to compare the effectiveness of static and contextual embeddings in enhancing entity recognition on Indonesian biomedical text. The experimental setup involved utilizing both static (Word2Vec) and contextual (IndoBERT) embeddings in conjunction with neural architectures (BiLSTM) along with Conditional Random Fields (CRF). The BiLSTM architecture was selected for its ability to capture bidirectional dependencies in language sequences. Specifically, four models: Word2Vec-BiLSTM, Word2Vec-BiLSTM-CRF, IndoBERT-BiLSTM, and IndoBERT-BiLSTM-CRF were evaluated to assess the impact of contextual representations and structured decoding. The models were trained on a manually annotated DetikHealth corpus, where specific medical entities such as diseases, symptoms, and drugs were labeled with the BIO-tagging scheme. Performance was subsequently evaluated based on standard metrics: precision, recall, and F1-score. Results indicate that IndoBERT’s contextual embeddings significantly outperform static Word2Vec features. The IndoBERT-BiLSTM-CRF model achieved the highest performance micro-F1 0.4330, macro-F1 0.3297, with the Disease entity reaching an F1-score of 0.5882. Combining contextual embeddings with CRF-based decoding enhances semantic understanding and boundary consistency, demonstrating superior performance for Indonesian biomedical NER. Future work should explore domain-adaptive pretraining and larger biomedical corpora to further improve contextual accuracy.
Domain Adaptation of Bert Models for Biomedical Entity Extraction from Indonesian Health News Norman, Maria Bernadette Chayeenee; Dewi, Ika Novita; Ignasius, Darnell
Jurnal ELTIKOM : Jurnal Teknik Elektro, Teknologi Informasi dan Komputer Vol. 10 No. 1 (2026)
Publisher : P3M Politeknik Negeri Banjarmasin

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31961/eltikom.v10i1.2116

Abstract

Health-related news articles play an increasingly important role in public health monitoring. However, their unstructured linguistic style complicates the automatic extraction of biomedical information. Indonesian health news shows high lexical variation by combining medical terms, colloquial expressions, borrowed Eng-lish words, and culturally specific symptom descriptions. This condition creates challenges for Named Entity Recognition (NER). To address the limited availability of domain-specific resources, this study compares four Transformer-based models, namely BERT, IndoBERT, RoBERTa, and BioBERT, for biomedical NER in Indone-sian health news. A new BIO-annotated dataset consisting of 272 manually labeled articles was constructed and validated, achieving strong inter-annotator agreement (Cohen’s Kappa = 0.88). To reduce data limita-tions, an additional 103 articles were automatically annotated using the best-performing model, RoBERTa, through a semi-supervised approach. All models were fine-tuned under identical settings and evaluated at both BIO and entity levels. The results show that RoBERTa achieves the highest weighted F1-score (0.9543). Howev-er, its macro F1-score (0.3873) indicates uneven performance across entity classes because of severe label im-balance, with non-entity tokens dominating the dataset. This finding highlights the importance of emphasizing macro-level evaluation to better reflect entity recognition performance. RoBERTa consistently outperforms the other models, which may be explained by its robust architecture and adaptability to diverse linguistic patterns. In contrast, BioBERT underperforms because of cross-lingual and domain mismatch, as it is pretrained on Eng-lish biomedical corpora and optimized for scientific text rather than journalistic language. The error analysis further identifies boundary inconsistencies and under-detection of low-frequency entities, especially in the drug and symptom categories.
Co-Authors Bernadette Chayeenee Norman , Maria Budi, Setyo Ika Novita Dewi Norman, Maria Bernadette Chayeenee Novita Dewi , Ika Ramadhan Rakhmat Sani