Optimizer-Based Performance Evaluation of Deep Learning Models for Loan Eligibility Prediction
Bibliographic record
Abstract
Reliable for loan eligibility prediction is necessary for banks in order to optimize the way they lend and reduce the risk of loan default. Traditional verification techniques are not keeping up with the increasing number of applications and customer demand for quick, reliable evaluations. Deep Neural Network (DNNs), Convolutional Neural Network (CNNs), and DenseNet are the three deep learning architectures that are studied and compared in our study. To classify loan eligibility outcomes, by training models using various Optimizers, such as Adam, Stochastic Gradient Descent, RMSprop, and Adagrad.To provide an adequate evaluation, we evaluate the models using a variety of performance metrices, which include accuracy, precision, recall, F1-score, and ROC-AUC. Our goal is to highlight that CNN models with Adam regularly outperform different strategies, obtained better generalization and predictive accuracy. On other hand, DenseNet models performed well on every metric. These results highlight how deep learning approaches can be used to automate loan eligibility predictions, providing a scalable method for improving loan risk evaluations and increasing operational effectiveness in the financial services industry.
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 | 0.006 | 0.011 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 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".