See veebileht kasutab küpsiseid kasutaja sessiooni andmete hoidmiseks. Veebilehe kasutamisega nõustute ETISe kasutustingimustega. Loe rohkem
Olen nõus
"Muu" projekt IKTRP-1
IKTRP-1 "Privaatsust säilitavad statistilised uuringud lingitud andmebaasidel (1.05.2013−31.08.2015)", Dan Bogdanov, Cybernetica AS.
Privaatsust säilitavad statistilised uuringud lingitud andmebaasidel
Privacy-preserving studies on linked databases
Teadus- ja arendusprojekt
ETIS klassifikaatorAlamvaldkondCERCS klassifikaatorFrascati Manual’i klassifikaatorProtsent
4. Loodusteadused ja tehnika4.6. ArvutiteadusedP170 Arvutiteadus, arvutusmeetodid, süsteemid, juhtimine (automaatjuhtimisteooria)1.1. Matemaatika ja arvutiteadus (matemaatika ja teised sellega seotud teadused: arvutiteadus ja sellega seotud teadused (ainult tarkvaraarendus, riistvara arendus kuulub tehnikavaldkonda)100,0
Cybernetica ASkoordinaator01.05.2013−31.08.2015
SA Archimedes
01.05.2013−31.08.2015256 047,00 EUR
256 047,00 EUR
IKT riiklik programm

Vaata ingliskeelset kokkuvõtet
We intend to develop a solution for a long-standing problem in organizational governance – the inability to combine confidential databases for performing statistical studies. Statistical studies assist the government in understanding long-term processes and trends in society and economy. While the information needed for a study can be obtained by performing a survey among a sample of people or institutions, this may constitute a duplication of effort with national registries and databases. There exist registries for education information (EHIS in Estonia), income information (databases of the Tax and Customs Board). Could these be used in statistical studies without compromising the privacy of involved individuals? We intend to develop better solutions for conducting studies based on linked databases. We will apply novel cryptographic techniques to ensure that information collected from individuals or linked databases can be analysed with the best possible privacy guarantees. Our approach will be based on the theory of secure multiparty computation, a cryptographic technique for performing computations on data so that the inputs remain private and only designated parties learn the outputs. We combine this technique with novel data collection and secure database technologies that allow several databases to be combined without disclosing the identities and sensitive data of the individuals in the database.