Comparative Evaluation on Effect of ELMo in Combination with Machine Learning, and Ensemble Models
Bibliographic record
Abstract
Communication using modern Internet technologies has revolutionized the ways that humans talk, write, and other types of exchange information.Despite all the advantages made available by information and communication technology, its applicability is still limited due to problems caused by personal attacks or pseudo-attacks, which are called toxic contents.These toxic contents may be in the form of texts including online chats, emails, speeches, or even images or clips on social media platforms.Since cyberbullying via the usage of toxic digital content on an individual may have severe consequences, it is important to design and implement various techniques to automatically detect cyberbullying from social media content using machine learning and deep learning approaches.During a cyberbullying detection process, word embedding techniques are used to represent the words for text analysis, typically in the form of a real-valued vector.These vectors encode the meaning of the word such that the words that are closer in the vector space are expected to be similar in meaning.The extracted embeddings are then used to identify if a digital input contains cyberbullying content.Feeding strong word representations to classification methods is an important issue.In this thesis, the effect of ELMo is evaluated against three other word embeddings namely, TF-IDF, Word2Vec, and BERT using various deep learning algorithms . . . . . . . . . . . .
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 | 0.007 | 0.009 |
| Meta-epidemiology (narrow) | 0.003 | 0.000 |
| Meta-epidemiology (broad) | 0.003 | 0.002 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 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".