Data transformation for privacy-preserving data mining
Bibliographic record
Abstract
The sharing of data is often beneficial in data mining applications. It has been proven useful to support both decision-making processes and to promote social goals. However, the sharing of data has also raised a number of ethical issues. Some such issues include those of privacy, data security, and intellectual property rights. In this thesis, we focus primarily on privacy issues in data mining, notably when data are shared before mining. Specifically, we consider some scenarios in which applications of association rule mining and data clustering require privacy safeguards. Addressing privacy preservation in such scenarios is complex. One must not only meet privacy requirements but also guarantee valid data mining results. This status indicates the pressing need for rethinking mechanisms to enforce privacy safeguards without losing the benefit of mining. These mechanisms can lead to new privacy control methods to convert a database into a new one in such a way as to preserve the main features of the original database for mining. In particular, we address the problem of transforming a database to be shared into a new one that conceals private information while preserving the general patterns and trends from the original database. To address this challenging problem, we propose a unified framework for privacy-preserving data mining that ensures that the mining process will not violate privacy up to a certain degree of security. The framework encompasses a family of privacy-preserving data transformation methods, a library of algorithms, retrieval facilities to speed up the transformation process, and a set of metrics to evaluate the effectiveness of the proposed algorithms, in terms of information loss, and to quantify how much private information has been disclosed. Our investigation concludes that privacy-preserving data mining is to some extent possible. We demonstrate empirically and theoretically the practicality and feasibility of achieving privacy preservation in data mining. Our experiments reveal that our framework is effective, meets privacy requirements, and guarantees valid data mining results while protecting sensitive information (e.g., sensitive knowledge and individuals' privacy).
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 | 0.018 | 0.046 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.004 |
| Bibliometrics | 0.003 | 0.006 |
| Science and technology studies | 0.002 | 0.004 |
| Scholarly communication | 0.005 | 0.008 |
| Open science | 0.003 | 0.006 |
| Research integrity | 0.002 | 0.006 |
| Insufficient payload (model declined to judge) | 0.002 | 0.002 |
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".