Method and system for data mining in high dimensional data spaces
First Claim
1. A computerized data mining method for analyzing a multitude of items in an n-dimensional space Dn, each described by n item features, said method using a mining function f with at least one control parameter Pi controlling the target of the data mining function, said method comprising a first step of selecting a transformation function T for reducing dimensions of said n-dimensional space by means of space-filling curves mapping said n-dimensional space to a m-dimensional space (m<
- n);
said method comprising a second step of determining a transformed control parameter PTi controlling the target of the data mining function in said m-dimensional space;
said method comprising a third step of applying said selected transformation function T on said multitude Dn of items to create a transformed multitude Dm of items; and
executing said mining function f controlled by said transformed control parameter PTi on said transformed multitude of items Dm.
2 Assignments
0 Petitions
Accused Products
Abstract
The proposed computerized method and system is adapted for analyzing a multitude of items in a high dimensional (n-dimensional) data space Dn each described by n item features. The method uses a mining function f with at least one control parameter Pi controlling the target of the data mining function.
A first step is selecting a transformation function T for reducing dimensions of the n-dimensional space by means of space-filling curves mapping said n-dimensional space to a m-dimensional space (m<n).
A second step is determining a transformed control parameter PTi controlling the target of the data mining function in the m-dimensional space.
A third step is applying the selected transformation function T on the multitude Dn of items to create a transformed multitude Dm of items and is executing the mining function f controlled by the transformed control parameter PTi on the transformed multitude of items Dm.
45 Citations
14 Claims
-
1. A computerized data mining method for analyzing a multitude of items in an n-dimensional space Dn, each described by n item features, said method using a mining function f with at least one control parameter Pi controlling the target of the data mining function,
said method comprising a first step of selecting a transformation function T for reducing dimensions of said n-dimensional space by means of space-filling curves mapping said n-dimensional space to a m-dimensional space (m< - n);
said method comprising a second step of determining a transformed control parameter PTi controlling the target of the data mining function in said m-dimensional space;
said method comprising a third step of applying said selected transformation function T on said multitude Dn of items to create a transformed multitude Dm of items; and
executing said mining function f controlled by said transformed control parameter PTi on said transformed multitude of items Dm. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14)
- n);
Specification