Data mining technique with position labeling
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 candidate database having a pool of candidate individuals, each candidate individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the training data and a position entering action to assert in dependence upon the training data, and the exiting rule identifying both a label to assert in dependence upon the training data and a position exiting action to assert in dependence upon the training data, each candidate individual further having associated therewith an indication of a respective fitness estimate; and
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, each trial applying the rules of the respective individual to the training data, each rule that asserts its action in response to the training data also asserting its label in response to the training data;
updates the fitness estimate associated with each of the individuals being tested in dependence upon both the training data and the actions and labels asserted by the rules in the respective individual in the battery of trials; and
selects individuals for discarding from the candidate pool in dependence upon predetermined criteria.
3 Assignments
0 Petitions
Accused Products
Abstract
Roughly described, individuals in both a training system and in a production system include a label field in their rule outputs. Positions entered by an individual are maintained in a status record for the individual, including the label output by the rule which triggered entry of that position. Rules that assert exiting or partial exiting of a position also output the label from the rule which triggered the assertion, and are effective only so far as matching positions exist or remain in the individual'"'"'s status record, including a matching label. Labels present in the status record also can be referenced in conditions of a rule. During evolution, a rule'"'"'s output label is subject to crossover and/or mutation just like the conditions and output assertions.
-
Citations
47 Claims
-
1. A computer-implemented data mining system, for use with a data mining training database containing training data, comprising:
-
a memory storing a candidate database having a pool of candidate individuals, each candidate individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the training data and a position entering action to assert in dependence upon the training data, and the exiting rule identifying both a label to assert in dependence upon the training data and a position exiting action to assert in dependence upon the training data, each candidate individual further having associated therewith an indication of a respective fitness estimate; and 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, each trial applying the rules of the respective individual to the training data, each rule that asserts its action in response to the training data also asserting its label in response to the training data; updates the fitness estimate associated with each of the individuals being tested in dependence upon both the training data and the actions and labels asserted by the rules in the respective individual in the battery of trials; and selects individuals for discarding from the candidate pool in dependence upon predetermined criteria. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20)
-
-
21. A computer-implemented data mining system, for use with a data mining training database containing training data, comprising:
-
a data processor; and a storage subsystem coupled to the data processor and having stored therein in a non-transitory manner a candidate database having a plurality of candidate individuals, each candidate individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the training data and a position entering action to assert in dependence upon the training data, and the exiting rule identifying both a label to assert in dependence upon the training data and a position exiting action to assert in dependence upon the training data, each candidate individual further having associated therewith an indication of a respective fitness estimate. - View Dependent Claims (22, 23, 24, 25)
-
-
26. A computer-implemented system, for use with an ordered sequence of input data, comprising:
-
a processor; and a storage subsystem coupled to the processor and having stored therein in a non-transitory manner a pool of at least one individual, each individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the input data and a position entering action to assert in dependence upon the input data, and the exiting rule identifying both a label to assert in dependence upon the input data and a position exiting action to assert in dependence upon the input data, each individual further having a status record associated therewith, wherein the processor, for each given individual in the pool; applies the rules of the given individual to the input data, each rule that asserts its action in response to the input data also asserting its label in response to the input data; and in response to activation of a given rule that asserts its action; outputs a signal toward a controlled system in dependence upon the action of the given rule, and updates a status record associated with the given individual in dependence upon both the action of the given rule and the label of the given rule. - View Dependent Claims (27, 28, 29, 30, 31, 32, 33, 34)
-
-
35. A method for data mining, for use with a data mining training database containing training data, comprising:
-
storing in a memory, a candidate database having a pool of candidate individuals, each candidate individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the training data and a position entering action to assert in dependence upon the training data, and the exiting rule identifying both a label to assert in dependence upon the training data and a position exiting action to assert in dependence upon the training data, each candidate individual further having associated therewith an indication of a respective fitness estimate; a candidate pool processor testing individuals from the candidate pool on the training data, each individual being tested undergoing a respective battery of at least one trial, each trial applying the rules of the respective individual to the training data, each rule that asserts its action in response to the training data also asserting its label in response to the training data; the candidate pool processor updating the fitness estimate associated with each of the individuals being tested in dependence upon both the training data and the actions and labels asserted by the rules in the respective individual in the battery of trials; and the candidate pool processor selecting individuals for discarding from the candidate pool in dependence upon predefined criteria. - View Dependent Claims (36, 37)
-
-
38. A method for data mining, for use with a data mining training database containing training data, comprising:
storing in a non-transitory manner in a storage subsystem coupled to a data processor, a candidate database having a plurality of candidate individuals, each candidate individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the training data and a position entering action to assert in dependence upon the training data, and the exiting rule identifying both a label to assert in dependence upon the training data and a position exiting action to assert in dependence upon the training data, each candidate individual further having associated therewith an indication of a respective fitness estimate.
-
39. A method for processing an ordered sequence of input data, comprising:
-
storing in a non-transitory manner in a storage subsystem coupled to a data processor, a pool of at least one individual, each individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the input data and a position entering action to assert in dependence upon the input data, and the exiting rule identifying both a label to assert in dependence upon the input data and a position exiting action to assert in dependence upon the input data, each individual further having a status record associated therewith; and for each given individual in the pool; applying the rules of the given individual to the input data, each rule that asserts its action in response to the input data also asserting its label in response to the input data, and in response to activation of a given rule that asserts its action; outputting a signal toward a controlled system in dependence upon the action of the given rule; and updating a status record associated with the given individual in dependence upon both the action of the given rule and the label of the given rule.
-
-
40. A computer-readable medium for implementing a data mining system and for use with a data mining training database containing training data,
the medium having stored thereon in a non-transitory manner a candidate database having a pool of candidate individuals, each candidate individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the training data and a position entering action to assert in dependence upon the training data, and the exiting rule identifying both a label to assert in dependence upon the training data and a position exiting action to assert in dependence upon the training data, each candidate individual further having associated therewith an indication of a respective fitness estimate; the medium further having stored thereon in a non-transitory manner a plurality of software code portions which define logic for implementing 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, each trial applying the rules of the respective individual to the training data, each rule that asserts its action in response to the training data also asserting its label in response to the training data, updates the fitness estimate associated with each of the individuals being tested in dependence upon both the training data and the actions and labels asserted by the rules in the respective individual in the battery of trials, and selects individuals for discarding from the candidate pool in dependence upon predetermined criteria. - View Dependent Claims (41, 42)
-
43. A computer-readable medium for implementing a data mining system and for use with a data mining training database containing training data,
the medium having stored therein in a non-transitory manner a candidate database having a plurality of candidate individuals, each candidate individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the training data and a position entering action to assert in dependence upon the training data, and the exiting rule identifying both a label to assert in dependence upon the training data and a position exiting action to assert in dependence upon the training data, each candidate individual further having associated therewith an indication of a respective fitness estimate.
-
44. A computer readable medium, for use with an ordered sequence of input data, the medium having stored therein in a non-transitory manner a pool of at least one individual, each individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the input data and a position entering action to assert in dependence upon the input data, and the exiting rule identifying both a label to assert in dependence upon the input data and a position exiting action to assert in dependence upon the input data, each individual further having a status record associated therewith,
the medium further having stored thereon in a non-transitory manner a plurality of software code portions which define logic for, for each given individual in the pool: -
applying the rules of the given individual to the input data, each rule that asserts its action in response to the input data also asserting its label in response to the input data; and in response to activation of a given rule that asserts its action; outputting a signal toward a controlled system in dependence upon the action of the given rule, and updating a status record associated with the given individual in dependence upon both the action of the given rule and the label of the given rule.
-
-
45. A computer-implemented data mining system, for use with a data mining training database containing training data, comprising:
-
storage means for storing a candidate database having a pool of candidate individuals, each candidate individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the training data and a position entering action to assert in dependence upon the training data, and the exiting rule identifying both a label to assert in dependence upon the training data and a position exiting action to assert in dependence upon the training data, each candidate individual further having associated therewith an indication of a respective fitness estimate; and candidate pool processor means for; testing individuals from the candidate pool on the training data, each individual being tested undergoing a respective battery of at least one trial, each trial applying the rules of the respective individual to the training data, each rule that asserts its action in response to the training data also asserting its label in response to the training data; updating the fitness estimate associated with each of the individuals being tested in dependence upon both the training data and the actions and labels asserted by the rules in the respective individual in the battery of trials; and selecting individuals for discarding from the candidate pool in dependence upon their updated fitness estimate.
-
-
46. A computer-implemented data mining system, for use with a data mining training database containing training data, comprising:
storage means for storing in a non-transitory manner a candidate database having a plurality of candidate individuals, each candidate individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the training data and a position entering action to assert in dependence upon the training data, and the exiting rule identifying both a label to assert in dependence upon the training data and a position exiting action to assert in dependence upon the training data, each candidate individual further having associated therewith an indication of a respective fitness estimate.
-
47. A computer-implemented data mining system, for use with an ordered sequence of input data, comprising:
-
storage means for storing in a non-transitory manner a pool of at least one individual, each individual identifying a respective set of at least two rules including an entering rule and an exiting rule, the entering rule identifying both a label to assert in dependence upon the input data and a position entering action to assert in dependence upon the input data, and the exiting rule identifying both a label to assert in dependence upon the input data and a position exiting action to assert in dependence upon the input data, each individual further having a status record associated therewith, and processing means for, for each given individual in the pool; applying the rules of the given individual to the input data, each rule that asserts its action in response to the input data also asserting its label in response to the input data; and in response to activation of a given rule that asserts its action; outputting a signal toward a controlled system in dependence upon the action of the given rule, and updating a status record associated with the given individual in dependence upon both the action of the given rule and the label of the given rule.
-
Specification