Bibliographic record
Abstract
With the emergence of Internet, it is now possible to connect and access sources of information and databases throughout the world. At the same time, this raises many questions regarding the privacy and the security of the data, in particular how to mine useful information while preserving the privacy of sensible and confidential data. Privacy-preserving data mining is a relatively new but rapidly growing field that studies how data mining algorithms affect the privacy of data and tries to find and analyze new algorithms that preserve this privacy. At first glance, it may seem that data mining and privacy have orthogonal goals, the first one being concerned with the discovery of useful knowledge from data whereas the second is concerned with the protection of data’s privacy. Historically, the interactions between privacy and data mining have been questioned and studied since more than a decade ago, but the name of the domain itself was coined more recently by two seminal papers attacking the subject from two very different perspectives (Agrawal & Srikant, 2000; Lindell & Pinkas, 2000). The first paper (Agrawal & Srikant, 2000) takes the approach of randomizing the data through the injection of noise, and then recovers from it by applying a reconstruction algorithm before a learning task (the induction of a decision tree) is carried out on the reconstructed dataset. The second paper (Lindell & Pinkas, 2000) adopts a cryptographic view of the problem and rephrases it within the general framework of secure multiparty computation. The outline of this chapter is the following. First, the area of privacy-preserving data mining is illustrated through three scenarios, before a classification of privacy- preserving algorithms is described and the three main approaches currently used are detailed. Finally, the future trends and challenges that await the domain are discussed before concluding.
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 | 0.025 | 0.051 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.002 |
| Bibliometrics | 0.003 | 0.006 |
| Science and technology studies | 0.003 | 0.018 |
| Scholarly communication | 0.012 | 0.017 |
| Open science | 0.003 | 0.009 |
| Research integrity | 0.007 | 0.011 |
| Insufficient payload (model declined to judge) | 0.009 | 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".