Cognitive e-KYC: A Convergence of Biometric Forensics and Digital Identity Validation
Bibliographic record
Abstract
e-KYC checks are of great importance for secure and efficient identity authentication in financial and regulatory services that are slow, error-prone and fraud-sensitive to traditional KYC methods. This study suggests an AI-driven e-KYC system that extracts important details such as AADHAAR number, name, date of birth, address, and integrates optical character detection (OCR) with Tesseract.js to achieve a processing time of 23 seconds per document with 90% accuracy of Aadhaar cards. The system uses deep learning-based facial recognition to compare it with 5% false negative rates in 1.5 seconds compared to live images and ID photos. Machine learning models recognize data inconsistencies to prevent fraud, but Uidai Aadhaar - API users details are detailed against official state records to improve reliability. Documentary security is guaranteed through ReportLab and Pypdf2 encryption. This adds 1.2 seconds of overhead per PDF with controlled file approval to meet data protection standards. End-to-end latency with stable internet remains under 10 seconds. This scalable and secure framework combines image improvements, document analysis, biometry, ML checks and API authentication to significantly improve the accuracy and recognition of fraud via traditional methods. Ideal for financial institutions, government agencies and digital platforms, it offers the latest automated AI improvement approach to reduce workloads, accelerate onboarding and check digital identity for manual reviews.
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 | 0.003 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.004 |
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".