Data mining technique with experience-layered gene pool
First Claim
1. A computer-implemented data mining system, for use with a data mining training database containing training data, comprising:
- a memory storing a database having a candidate pool comprising a set of individuals, each candidate individual identifying a plurality of conditions and at least one corresponding consequence to be asserted in dependence upon the conditions, each candidate individual further having associated therewith an indication of a respective fitness estimate, and an indication of a respective testing experience level;
a candidate pool processor which;
tests individuals from the candidate pool on the training data, each individual being tested undergoing a respective battery of at least one trial and thereby increasing the individual'"'"'s testing experience level, each trial applying the conditions of the respective individual to the training data to assert a consequence,updates the fitness estimate associated with each of the individuals being tested in dependence upon the training data and consequences asserted by the individuals in the testing on the battery of trials;
wherein the candidate pool processor comprises a competition module which;
assigns each individual into one of a plurality of testing experience groups in dependence upon the individual'"'"'s testing experience level, andselects an individual from each of at least two of the testing experience groups for discarding from the candidate pool in dependence upon both their testing experience group and a predetermined criteria.
4 Assignments
0 Petitions
Accused Products
Abstract
Roughly described, a computer-implemented evolutionary data mining system includes a memory storing a candidate gene database in which each candidate individual has a respective fitness estimate; a gene pool processor which tests individuals from the candidate gene pool on training data and updates the fitness estimate associated with the individuals in dependence upon the tests; and a gene harvesting module providing for deployment selected ones of the individuals from the gene pool, wherein the gene pool processor includes a competition module which selects individuals for discarding from the gene pool in dependence upon both their updated fitness estimate and their testing experience level. Preferably the gene database has an elitist pool containing multiple experience layers, and the competition module causes individuals to compete only with other individuals in their same experience layer.
37 Citations
58 Claims
-
1. A computer-implemented data mining system, for use with a data mining training database containing training data, comprising:
-
a memory storing a database having a candidate pool comprising a set of individuals, each candidate individual identifying a plurality of conditions and at least one corresponding consequence to be asserted in dependence upon the conditions, each candidate individual further having associated therewith an indication of a respective fitness estimate, and an indication of a respective testing experience level; a candidate pool processor which; tests individuals from the candidate pool on the training data, each individual being tested undergoing a respective battery of at least one trial and thereby increasing the individual'"'"'s testing experience level, each trial applying the conditions of the respective individual to the training data to assert a consequence, updates the fitness estimate associated with each of the individuals being tested in dependence upon the training data and consequences asserted by the individuals in the testing on the battery of trials; wherein the candidate pool processor comprises a competition module which; assigns each individual into one of a plurality of testing experience groups in dependence upon the individual'"'"'s testing experience level, and selects an individual from each of at least two of the testing experience groups for discarding from the candidate pool in dependence upon both their testing experience group and a predetermined criteria. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
-
10. A computer-implemented data mining system, comprising a server with respect to a collection of at least one client device, the server comprising:
-
a memory storing a database having a candidate pool comprising a set of candidate individuals, each candidate individual identifying a plurality of conditions and at least one corresponding consequence to be asserted in dependence upon the conditions, each candidate individual further having associated therewith an indication of a respective fitness estimate, and an indication of a respective testing experience level; a candidate pool processor; and a candidate harvesting module providing for deployment of selected ones of the individuals from the candidate pool, wherein the candidate pool processor comprises; a client delegation module which delegates testing of individuals from the candidate pool to the at least one client device; and a competition module which; receives back individuals from the at least one client device after having undergone their respective testing batteries, updates the fitness estimate associated with each individual received back from the client devices in dependence upon results of its respective testing batteries, assigns each individual into one of a plurality of testing experience groups in dependence upon the individual'"'"'s testing experience level, and selects an individual from each of at least two of the testing experience groups for discarding from the candidate pool in dependence upon both their testing experience group and a predetermined criteria. - View Dependent Claims (11, 12, 13, 14, 15, 16, 17)
-
-
18. A computer-implemented data mining system, for use with a data mining training database containing training data, comprising a client with respect to a server device, the client comprising:
-
a memory storing a database having a client candidate pool comprising a set of individuals, each candidate individual identifying a plurality of conditions and at least one corresponding consequence to be asserted in dependence upon the conditions, each candidate individual further having associated therewith an indication of a respective client-centric fitness estimate, and an indication of a respective testing experience level; a module which receives individuals from the server device to be tested and inserts them into the client candidate pool; a testing module that; tests individuals from the client candidate pool on the training data, each individual being tested by the client undergoing a respective battery of at least one trial, each trial by the client applying the conditions of the respective individual to the training data to assert a consequence, and updates the client-centric fitness estimate associated with each of the individuals being tested by the client in dependence upon the training data and consequences asserted by the individuals in the testing on the battery of trials; a candidate pool processor having a competition module that; assigns each individual from the client candidate pool into one of a plurality of testing experience groups in dependence upon the individual'"'"'s testing experience level, and selects individuals for discarding from the client candidate pool in dependence upon both their testing experience level and other predetermined criteria; and a candidate harvesting module that forwards selected ones of the individuals from the client candidate pool to the server device. - View Dependent Claims (19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29)
-
-
30. A computer-implemented data mining method, for use with a data mining training database containing training data, comprising the steps of:
-
storing in a database a candidate pool comprising a set of candidate individuals, each candidate individual identifying a plurality of conditions and at least one corresponding consequence to be asserted in dependence upon the conditions, each candidate individual further having associated therewith an indication of a respective fitness estimate, and an indication of a respective testing experience level; testing individuals from the candidate pool on the training data, each individual being tested undergoing a respective battery of at least one trial and thereby increasing the individual'"'"'s testing experience level, each trial applying the conditions of the respective individual to the training data to assert a consequence; updating the fitness estimate associated with each of the individuals being tested in dependence upon the training data and consequences asserted by the individuals in the testing on the battery of trials; assigning each individual into one of a plurality of testing experience groups in dependence upon the individual'"'"'s testing experience level; selecting an individual from each of at least two of the testing experience groups for discarding from the candidate pool in dependence upon both their testing experience group and a predetermined criteria. - View Dependent Claims (31, 32, 33, 34, 35, 36, 37)
-
-
38. A computer-implemented data mining method performed by a server device, the method comprising the steps of:
-
storing in a database a candidate pool comprising a set of candidate individuals, each candidate individual identifying a plurality of conditions and at least one corresponding consequence to be asserted in dependence upon the conditions, each candidate individual further having associated therewith an indication of a respective fitness estimate, and an indication of a respective testing experience level; delegating testing of individuals from the candidate pool to at least one client device; receiving back individuals returned from the at least one client device after having undergone their respective testing batteries, each returned individual having associated therewith an indication of the results of its respective testing battery; updating the fitness estimate associated with each returned individual in dependence upon the results of its respective testing batteries; assigning each candidate individual into one of a plurality of testing experience groups in dependence upon the individual'"'"'s testing experience level; selecting an individual from at least two of the testing experience groups for discarding from the candidate pool in dependence upon both their testing experience group and fitness estimate; and deploying of selected ones of the individuals from the candidate pool. - View Dependent Claims (39, 40, 41, 42, 43, 44)
-
-
45. A computer-implemented data mining method performed by a client device, for use with a data mining training database containing training data, the method comprising:
-
storing in a database a client candidate pool comprising a set of candidate individuals, each candidate individual identifying a plurality of conditions and at least one corresponding consequence to be asserted in dependence upon the conditions, each candidate individual further having associated therewith an indication of a respective client-centric fitness estimate, and an indication of a respective testing experience level; receiving from a server device individuals to be tested and inserting them into the client candidate pool; testing individuals from the client candidate pool on the training data, each individual being tested by the client undergoing a respective battery of at least one trial, each trial by the client applying the conditions of the respective individual to the training data to assert a consequence; updating the client-centric fitness estimate associated with each of the individuals being tested by the client in dependence upon the training data and consequences asserted by the individuals in the testing on the battery of trials; assigning each individual from the client candidate pool into one of a plurality of testing experience groups in dependence upon the individual'"'"'s testing experience level; selecting individuals for discarding from the client candidate pool in dependence upon both their testing experience level and other predetermined criteria; and forwarding selected ones of the individuals from the client candidate pool to the server device. - View Dependent Claims (46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57)
-
-
58. A computer-implemented data mining method, for use with a data mining training database containing training data, comprising the steps of:
-
storing in a database a candidate pool comprising a set of candidate individuals, each candidate individual identifying a plurality of conditions and at least one corresponding consequence to be asserted in dependence upon the conditions, each candidate individual further having associated therewith an indication of a respective fitness estimate, and an indication of a respective testing experience level; defining a plurality of testing experience level groups for the candidate pool, each of the groups being assigned a respective distinct range of testing experience levels; testing individuals from the candidate pool on the training data, each individual being tested undergoing a respective battery of at least one trial and thereby increasing the individual'"'"'s testing experience level, each trial applying the conditions of the respective individual to the training data to assert a consequence; updating the fitness estimate associated with each of the individuals being tested in dependence upon the training data and consequences asserted by the individuals in the testing on the battery of trials; selecting an individual from each of at least two of the testing experience level groups for discarding from the candidate pool in dependence upon both their testing experience level group and a predetermined criteria; and deploying selected ones of the individuals from the candidate pool.
-
Specification