A Scalable Architecture for Automated Data Classification and Sensitive Information Discovery Using Artificial Intelligence
Bibliographic record
Abstract
The continuous expansion of enterprise data across cloud computing platforms, distributed storage systems, and digital communication networks has significantly increased the complexity of managing and securing sensitive information. Traditional rule-based and manual data classification techniques are often inadequate for handling large-scale heterogeneous datasets due to limited scalability, low contextual awareness, and high operational overhead. With the increasing complexity of enterprise data governance, privacy protection, and compliance with cybersecurity regulations, this paper presents an AI-powered, scalable solution for automated data classification and sensitive information discovery. The proposed solution combines machine learning, deep learning, Natural Language Processing (NLP) and transformer-based models to automatically classify enterprise structured, semi-structured and unstructured data. The architecture features several functional components, such as data ingestion, data preprocessing, classification by AI, discovery of sensitive data, compliance management, and secure data storage. By using advanced NLP and Named Entity Recognition (NER) techniques, entities that need to be kept confidential are accurately identified, including personally identifiable information (PII), healthcare records, financial data, and organizational secrets. Cloud-native distributed processing and scalable monitoring frameworks further amplify processing efficiency, flexibility and real-time data governance features. The evaluation results from experiments show that the proposed architecture using AI outperforms the traditional rule-based architecture for classification accuracy, sensitive data detection performance, scalability, and operational efficiency. The framework also features automated governance and auditing to help ensure that all regulations are met, including GDPR, HIPAA, and CCPA. In conclusion, the proposed architecture offers a secure and intelligent way to manage enterprise data in today's digital landscape.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.010 | 0.008 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.006 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".