Text Augmentation Approaches to Enhance Traditional Machine Learning Performance for SDGs Classification of Indonesian News Articles
Bibliographic record
Abstract
The Sustainable Development Goals (SDGs) require effective monitoring, yet detecting SDG-related content in Indonesian texts is difficult due to limited resources, Code-Mixing, and the multilabel nature of the task.One article may correspond to several goals, creating imbalance and inter-label dependencies that complicate classification.This study applies multilabel classification for Indonesian SDG news using Naï ve Bayes, Logistic Regression, Support Vector Machine, and Random Forest with TF-IDF features and 5-Fold Cross Validation.However, these approaches showed limited performance.To improve results, four data augmentation strategies were explored for oversampling: Code-Mixing with Back Translation, Code-Mixing with Paraphrased Back-Translation, Simple Back Translation, and Paraphrased Back Translation.From 4,195 original articles on Universitas Gadjah Mada websites, 5,105 augmented samples were generated, producing 9,300 documents.Experiments show that augmentation reduces imbalance and enhances classification.SVM and RF achieved the best results, with F1-Scores above 0.93 and Hamming Loss between 0.028 and 0.067, while LR was competitive with higher efficiency.Among augmentation methods, the most effective were Code-Mixing with Back Translation and Simple Back Translation without paraphrasing.Overall, this study demonstrates that augmentation can significantly improve traditional and lightweight classifiers, offering a practical and resource-efficient alternative for SDG multilabel classification in Indonesian news and other comparable low-resource text environments.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".