Data retrieval in pattern recognition systems
Methods, apparatuses and systems directed to pattern identification and pattern recognition. In some particular implementations, the invention provides a flexible pattern recognition platform including pattern recognition engines that can be dynamically adjusted to implement specific pattern recognition configurations for individual pattern recognition applications. In some implementations, the present invention also provides for a partition configuration where knowledge elements can be grouped and pattern recognition operations can be individually configured and arranged to allow for multi-level pattern recognition schemes.
1. A system, comprising memory and logic operatively coupled to the memory and configured to:
receive a first vector;
in a learning operation, associate the first vector with a first knowledge element of a knowledge base;
store an association between the first knowledge element and remote data stored in a remote database separate from the knowledge base;
receive a second vector;
in a recognition operation, match the second vector to the first knowledge element; and
in response to the second vector matching the first knowledge element, retrieve the remote data from the remote database based on the association between the remote data and the first knowledge element.
2. The system of claim 1 , wherein the logic is configured to retrieve the remote data using an identifier of the first knowledge element.
3. The system of claim 1 , wherein the logic is configured to retrieve the remote data using information associated with the first knowledge element but not processed by the knowledge base.
4. The system of claim 3 , wherein the information include or reference one or more of: the first vector, training data from which the first vector was derived, a knowledge base partition for a subsequent recognition operation, all or a portion of an input vector for a subsequent recognition operation, or a weighting for a subsequent recognition operation.
5. The system of claim 1 , wherein the logic is configured to generate the first knowledge element using the first vector.
6. The system of claim 1 , wherein the remote data are training data from which the first vector was derived.
7. The system of claim 6 , wherein the logic is configured to retrieve the training data in response to a request generated by an auditing application.
8. The system of claim 7 , wherein the logic is further configured to present a representation of the training data in a user interface of the auditing application.
9. The system of claim 1 , wherein the first knowledge element is one of a plurality of knowledge elements of the knowledge base, and wherein the logic is further configured to maintain a count for each of the knowledge elements representing a number of matched vectors corresponding to that knowledge element.
10. The system of claim 1 , wherein the first knowledge element is one of a plurality of knowledge elements of the knowledge base, and wherein the logic is further configured to maintain a count for each of the knowledge elements representing a number of times a user-defined rule was satisfied for that knowledge element.
11. The system of claim 1 , wherein the logic is configured to match the second vector to the first knowledge element using one of a radial basis function, a K-nearest-neighbors function, a Bayesian function, a linear function, or an artificial neural network.
12. The system of claim 1 , wherein the logic and memory are implemented as one or more computing devices, one or more programmable logic devices, one or more field-programmable gate arrays, or one or more application-specific integrated circuits.
13. A computer-implemented method, comprising:
receiving a first vector;
in a learning operation, associating the first vector with a first knowledge element of a knowledge base;
storing an association between the first knowledge element and remote data stored in a remote database separate from the knowledge base;
receiving a second vector;
in a recognition operation, matching the second vector to the first knowledge element; and
in response to the second vector matching the first knowledge element, retrieving the remote data from the remote database based on the association between the remote data and the first knowledge element.
14. The method of claim 13 , wherein the remote data are retrieved using an identifier of the first knowledge element.
15. The method of claim 13 , wherein the remote data are retrieved using information associated with the first knowledge element but not processed by the knowledge base.
16. The method of claim 15 , wherein the information include or reference one or more of: the first vector, training data from which the first vector was derived, a knowledge base partition for a subsequent recognition operation, all or a portion of an input vector for a subsequent recognition operation, or a weighting for a subsequent recognition operation.
17. The method of claim 13 , wherein associating the first vector with the first knowledge element includes generating the first knowledge element using the first vector.
18. The method of claim 13 , wherein the remote data are training data from which the first vector was derived.
19. The method of claim 18 , wherein the training data are retrieved in response to a request generated by an auditing application.
20. The method of claim 19 , further comprising presenting a representation of the training data in a user interface of the auditing application.
21. The method of claim 13 , wherein the first knowledge element is one of a plurality of knowledge elements of the knowledge base, the method further comprising maintaining a count for each of the knowledge elements representing a number of matched vectors corresponding to that knowledge element.
22. The method of claim 13 , wherein the first knowledge element is one of a plurality of knowledge elements of the knowledge base, the method further comprising maintaining a count for each of the knowledge elements representing a number of times a user-defined rule was satisfied for that knowledge element.
23. The method of claim 13 , wherein the second vector is matched to the first knowledge element using one of a radial basis function, a K-nearest-neighbors function, a Bayesian function, a linear function, or an artificial neural network.
24. A computer program product, comprising one or more non-transitory computer-readable media having computer program instructions stored therein, the computer program instructions being configured such that, when executed by one or more computing devices, the computer program instructions cause the one or more computing devices to:
receive a first vector;
in a learning operation, associate the first vector with a first knowledge element of a knowledge base;
store an association between the first knowledge element and remote data stored in a remote database separate from the knowledge base;
receive a second vector;
in a recognition operation, match the second vector to the first knowledge element; and
in response to the second vector matching the first knowledge element, retrieve the remote data from the remote database based on the association between the remote data and the first knowledge element.