Support Vector Machines Optimisation for Face Recognition Using Sparrow Search Algorithm
Bibliographic record
Abstract
In the realm of face recognition utilising Support Vector Machines (SVM), the adaptivity of the penalty parameter c and the kernel function g is often found lacking, leading to suboptimal recognition rates.To address this issue, an approach harnessing the Sparrow Search Algorithm (SSA) for SVM parameter optimisation has been proposed.Traditional methods such as grid and random search, alongside other swarm intelligence optimisation algorithms like Particle Swarm Algorithm (PSO) and Differential Evolutionary Algorithm (DE), were surpassed by the capabilities of the SSA in numerous applications.Crossvalidation (CV) was employed, with the SVM model training recognition accuracy serving as the SSA fitness value.Upon achieving optimal fitness values, the best combination of hyperparameters was ascertained.The overarching aim was to deploy the SSA for global optimisation of SVM's penalty parameters and kernel function, ensuring the derivation of the globally optimal solution for the ultimate classifier model in face recognition tasks.An empirical analysis conducted on the ORL standard face database revealed that the proposed method outperformed the PSO, DE, Gray Wolf Algorithm (GWO), and Enhanced Gray Wolf Algorithm (EGWO), registering an average accuracy of 95.1%.This starkly contrasts with the 81.9% accuracy of the traditional SVM.Such results demonstrate the method's efficacy in enhancing recognition performance, offering a novel avenue to elevate the accuracy of conventional SVM-based face recognition.
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 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 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".