KI-Glossar
Das vollständige Wörterbuch der Künstlichen Intelligenz
Private Set Intersection
Cryptographic protocol allowing two parties to discover the intersection of their private data sets without revealing other information. This method is particularly useful for collaboration between organizations while preserving confidentiality.
Secure Enclaves
Hardware-level isolated and secure memory area that ensures the confidentiality and integrity of the code and data executed there. Secure enclaves protect against unauthorized access even by the operating system or hypervisor.
k-Anonymity
Privacy protection model ensuring that each record in an anonymized dataset is indistinguishable from at least k-1 other records. This approach prevents re-identification through cross-referencing of quasi-identifiers.
l-Diversity
Extension of the k-anonymity model ensuring that for each group of quasi-identical records, there are at least l distinct values for sensitive attributes. This measure strengthens protection against attribute inference attacks.
t-Closeness
Privacy criterion requiring that the distribution of sensitive values in each equivalence group be close to the overall distribution in the dataset. This approach prevents distribution inference attacks in anonymized data.
Secure Function Evaluation
Protocol allowing multiple parties to collaborate to evaluate a function on their combined private inputs without revealing these inputs individually. This technology enables collaborative analysis while preserving the confidentiality of participants' data.
Distributed Ledger Privacy
Set of cryptographic techniques applied to distributed ledgers to ensure transaction confidentiality while maintaining transparency and auditability. These mechanisms include blind signatures, commitments, and zero-knowledge proofs.