HG-SSA-ChurnNet: A Hybrid Gradient-Guided Salp Swarm Optimized Deep Learning Framework for Telecom Analytics
Bibliographic record
Abstract
In this era, the precise analysis of user behavior and network performance in tele communications is essential for demonstrating effective service quality and reducing customer churn. Specifically, when compared to deep learningbased models like ChurnNet, which provided good prediction accuracy for telecom data, their performance is very sensitive to the choice of hyper-parameters that are usually selected through soft or heuristic trial and error methods. To address this limitation, this research proposes a Hybrid Gradient Guided Salp Swarm Algorithm with ChurnNet (HG-SSAChurnNet) for churn prediction and modeling of DNNs. In particular, the proposed HG-SSA-ChurnNet framework incorporates chaotic initialization for search diversity improvement, validation-loss gradient-based guiding for fast convergence and diversity-preserving follower dynamics that helps prevent early stagnation. Therefore, these mechanisms allow the optimizer to effectively explore the ChurnNet hyperparameter space, which includes learning rate, convolutional filters size, kernel sizes, dropout probability, mini-batch size, and optimizer type. Subsequently, the optimization is determined by a telecom-aware multi-objective fitness function that includes accuracy, F1-score and training time. Thus, the experimental analysis demonstrates that the proposed HG-SSA-ChurnNet outperforms the existing ChurnNet model with$\mathbf{8 1. 9 6 \%}$accuracy and$\mathbf{7 8. 8 4 \%}$F1-score.
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 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".