Split leverage: attacking the confidentiality of linked databases by partitioning
Bibliographic record
Abstract
This article considers the risk of disclosure in linked databases when statistical analysis of micro-data is permitted. The risk of disclosure needs to be balanced against the utility of the linked data. The current work specifically considers the disclosure risks in permitting regression analysis to be performed on linked data. A new attack based on partitioning of the database is presented. References Anton, H., Rorres, C. Elementary Linear Algebra, 10th edition. John Wiley and Sons Inc, Hoboken, NJ, 2010. Chipperfield, J. O., Yu, F., Gare, M. Providing access to microdata for statistical purposes: Experiences of the Australian Bureau of Statistics with remote analysis servers. Symposium 2011, Catalogue no. 11-522-XCB, Statistics Canada, pp.187–194, http://publications.gc.ca/collections/collection_2013/statcan/11-522-x/CS11-522-2011-eng.pdf, 2011. Cox, L. Confidentiality Issues For Statistical Database Query Systems. Invited Paper for Joint UNECE/Eurostat Seminar on Integrated Statistical Information Systems and Related Matters, Geneva Switzerland, http://www.unece.org/stats/documents/ces/sem, 2002. Dwork, C., McSherry, F., Nissim, K., Smith, A. Calibrating noise to sensitivity in private data analysis. Proceedings of the 3rd Theory of Cryptography Conference, LNCS 3876, pp. 265–284, 2006. doi:10.1007/11681878_14 Duncan, G. T., Elliott, M., Salazar-Gonzalez, J.-J. Statistical Confidentiality: Principles and Practice. Springer, NY, 2012. doi:10.1007/978-1-4419-7802-8 Gomatam, S., Karr, A., Reiter, J., Sanil, A. Data dissemination and disclosure limitation in a world without microdata: A risk-utility framework for remote access systems. Statistical Science 20(2), pp. 163–177, 2005. doi:10.1214/088342305000000043 Hundepool, A., Domingo-Ferrer, J., Franconi, L., Giessing, S., Norholdt, E., Spicer, K., de Wolf, P.-P. Statistical Disclosure Control. Wiley, UK, 2012. O'Keefe, C., Chipperfield, J. A Summary of Attack Methods and Confidentiality Protection Measures for Fully Automated Remote Analysis Systems. International Statistical Review 81(3), pp. 426–455, 2013. doi:10.1111/insr.12021 O'Keefe, C., Good, N. Regression output from a remote analysis server. Data and Knowledge Engineering 68(11), pp. 1175–1186, 2009. doi:10.1016/j.datak.2009.06.009 Reiter, J. P. Model diagnostics for remote-access regression servers. Statistics and Computing 13(4), pp. 371–380, 2003. doi:10.1023/A:1025623108012 Reiter, J. P., Kohnen, C. N. Categorical data regression diagnostics for remote servers. Journal of Statistical Computation and Simulation 75(11), pp. 889–903, 2005. doi:10.1080/00949650412331299184 Reznek, A. P. Recent confidentiality research related to access to enterprise microdata. Comparative Analysis of Enterprise Microdata Conference Chicago, IL, USA, http://www.oecd.org/std/37503027.pdf, 2006. Ritchie, F. Disclosure Controls for Regression Outputs. London: Mimeo, Office of National Statistics, London, http://www.wiserd.ac.uk/files/7913/6543/6668/WISERD_WDR_006.pdf, 2006. Sparks, R., Carter, C., Donnelly, J., Duncan, J., O'Keefe, C., Ryan, L. A framework for performing statistical analyses of unit record health data without violating either privacy or confidentiality of individuals. Proceedings of the 55th Session of the International Statistical Institute, Sydney, 2005. Sparks, R., Carter, C., Donnelly, J. B., O'Keefe, C., Duncan, J., Keighley, T., McAullay, D. Remote access methods for exploratory data analysis and statistical modelling: Privacy-Preserving Analytics. Computer Methods and Programs in Biomedicine 91(3), pp. 208–222, 2008. doi:10.1016/j.cmpb.2008.04.001 Wetherill, G. B. Regression Analysis with Applications. Chapman and Hall Ltd, London, 1986. Wolfram Research, Inc. Mathematica Version 8.0. Wolfram Research, Inc., Champaign, IL, USA, 2010.
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 | 0.022 | 0.069 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.003 | 0.004 |
| Science and technology studies | 0.003 | 0.005 |
| Scholarly communication | 0.007 | 0.023 |
| Open science | 0.004 | 0.022 |
| Research integrity | 0.003 | 0.006 |
| Insufficient payload (model declined to judge) | 0.003 | 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".