Phishing URL Detection Using Deep Learning: A Resilient Approach to Mitigating Emerging Cybersecurity Threats
Bibliographic record
Abstract
Phishing via malicious URLs remains a significant cybersecurity threat, exacerbated by the increasing dependence on digital platforms for communication, transactions, and data exchange.The ability to accurately distinguish between legitimate and phishing URLs is critical for safeguarding sensitive information and mitigating cyber threats.This study proposes a deep learning-based phishing URL detection model that processes raw URL input without requiring manual feature engineering.The model integrates char-acter-level embeddings with a hybrid parallel CNN-BiGRU architecture, leveraging Parallel CNN layers for local pattern extraction and BiGRU for capturing sequential dependencies in URL structures.The experimental results demonstrate that the proposed model achieves 98.46% accuracy, an AUC curve score of 99.62%, precision 98.45%, and recall 98.46%, along with a F1-score 98.45%.The hybrid architecture outperforms the utilization of individual CNNs since it combines parallel convolutional layers for local features as well as BiGRU for sequential relationships to offer more balanced and global performance on all of the measure metrics surpassing the performance of existing phishing detection frameworks.These findings underscore the effectiveness of combining convolutional and recurrent neural networks to enhance phishing detection capabilities.The study contributes to advancing cybersecurity defenses by providing an efficient, reliable, and scalable deep learning-based phishing detection framework capable of adapting to evolving phishing tactics.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".