FuzzyPsi
Models & Types · AiDotNet.FederatedLearning.PSI
Implements approximate entity matching using multiple similarity strategies.
Inheritance: PsiBase → FuzzyPsi
For Beginners
In real-world data, the same entity may have slightly different identifiers across organizations. “John Smith” at one hospital might be “Jon Smith” or “SMITH, JOHN” at another. Fuzzy matching finds these approximate matches.
How It Works
This class provides a unified PSI wrapper around fuzzy matching that first applies approximate matching to find candidate pairs, then runs the underlying PSI protocol on the matched identifiers.
Constructors
| Constructor | Summary |
|---|---|
FuzzyPsi(IPrivateSetIntersection innerProtocol) |
Initializes a new instance of FuzzyPsi wrapping an inner PSI protocol. |
Properties
| Property | Summary |
|---|---|
string ProtocolName { get; } |