Bibliographic record
Abstract
Cyberbullying can be defined as a form of bullying that occurs across social media platforms using electronic messages [1].With the widespread of Internet, there is a tremendous increase in the number of social media users who can access various online platforms.These platforms provide a ground for the cyberbullies to engage in bullying activities by spreading online gossips, and posting hateful comments thereby targeting a victim group.The impact of consequences on victims is unpredictable given such attacks can cause serious physical and psychological harm.It is important to implement measures to identify and prevent cyberbullying.State-of-the-art technologies such as Machine learning, Natural language processing and Deep learning can be used for identifying and studying the patterns required to develop models that can detect cyberbullying.This dissertation proposes three different approaches and five models based on these technologies to identify cyberbullying using a newly generated email dataset.This email dataset was derived from multiple online platforms.Our initial approach consists in using a traditional supervised machine learning, which is a classic method.Our second approach is based on DistilBERT [2].Our last approach employs an ensemble technique that combines (or equivalently, 'stacks') these base models to study the impact on performance.Our initial approach led to the implementation of two SVM (Support Vector Machine) models [3]: one using TF-IDF (Term Frequency-Inverse Document Frequency) First and foremost, with my most sincere gratitude, I would like to acknowledge my supervisor Dr. Wei Shi for her dedication, patience, encouragement, and support.Without her guidance, knowledge and expertise, this dissertation would have not been possible.She has been with me throughout the research process and helped me academically and personally.I am
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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".