Enhancing Cyberbullying Detection on Indonesian Twitter: Leveraging FastText for Feature Expansion and Hybrid Approach Applying CNN and BiLSTM
Bibliographic record
Abstract
Cyberbullying, characterized by the transmission of threatening, intimidating, and derogatory messages via digital platforms such as Twitter, is a pervasive issue.Given the volume of approximately 867 million daily tweets, the potential scale of cyberbullying incidents is immense, underscoring the necessity for automated detection systems for such messages.However, the context-sensitive nature of tweets can pose challenges to understanding message content, particularly in languages like Indonesian with potential for significant vocabulary discrepancies.This study aims to enhance cyberbullying detection by employing feature expansion using FastText, thereby addressing vocabulary-related comprehension issues in Indonesian-language tweets.Furthermore, text classification is performed using a Hybrid Deep Learning approach, integrating Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM).This hybrid model leverages the strengths of both techniques, capturing local patterns and long-range dependencies within the data.The objective of this research is to evaluate the performance yielded by the application of FastText-enhanced feature expansion and Hybrid Deep Learning to an Indonesian Twitter dataset.This focus is motivated by the high accuracy of Hybrid Deep Learning for Twitter datasets in other languages, and the limited application of such methods to Indonesian-language datasets, which predominantly use supervised learning or deep learning.Analysis of 29,085 datasets demonstrated that the combined implementation of Hybrid Deep Learning and FastText-enhanced feature expansion achieved the highest accuracy, with CNN-BiLSTM and BiLSTM-CNN scoring 80.55% and 80.35% respectively.These findings validate the significant accuracy boost provided by FastText when integrated with Hybrid Deep Learning.It is anticipated that the outcomes of this study will facilitate the accurate identification and removal of cyberbullying tweets, thereby contributing to a safer digital communication environment on Twitter.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 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) | 0.002 | 0.001 |
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".