System and method for machine learning management
According to one aspect, a method for machine learning management is provided. In one embodiment, the method includes receiving a first segment of text data, identifying data features corresponding to a sequence of characters in the first segment of text data, and generating predictive annotations to the sequence of characters based at least in part on the identified data features. The method can also include identifying inaccurate annotations generated according to the predictive annotations, correcting the identified inaccurate annotations, generating one or more sets of model training data incorporating the corrected annotations, and monitoring progress of annotations made to a second segment of text data associated with the first segment of text data by a plurality of collaborating users of a plurality of managed computers.
1. A computer-implemented method, comprising:
receiving a first segment of text data,
identifying data features corresponding to a sequence of characters in the first segment of text data,
generating predictive annotations to the sequence of characters based at least in part on the identified data features,
identifying inaccurate annotations generated according to the predictive annotations,
correcting the identified inaccurate annotations,
generating at least one set of model training data incorporating the corrected annotations, and
monitoring progress of annotations made to a second segment of text data associated with the first segment of text data by a plurality of collaborating users of a plurality of managed computers, the monitoring including determining, based at least in part on a training descriptor corresponding to the second segment of text data, a state of completion of annotations made to the second segment of text data by a particular one of the plurality of collaborating users, wherein the training descriptor identifies types of annotations in the at least one set of model training data.
2. The method of claim 1 , wherein at least one of receiving the first segment of text data, identifying the data features, generating the predictive annotations, identifying the inaccurate annotations, correcting the identified inaccurate annotations, generating the at least one set of model training data, and monitoring the progress of annotations is performed in response to receiving user input data that is received from a user via a user interface that is operatively coupled to the processing unit.
3. The method of claim 1 , wherein the state of completion of annotations made to the second segment of text data by the particular one of the plurality of collaborating users includes a state of incomplete, in-process annotation.
4. The method of claim 1 , wherein the training descriptor further identifies resource data corresponding to the at least one set of model training data, the resource data comprising a lexicon of associations between predefined sequences of characters and predefined labels.
5. The method of claim 1 , wherein the types of annotations identified by the training descriptor comprise at least one of categories and parts of speech associated with the sequence of characters.
6. The method of claim 1 , wherein the training descriptor further identifies types of features in the identified data features and a relative weight assigned to each of the types of features in the identified data features.
7. The method of claim 1 , further comprising assigning the training descriptor to the generated set of model training data.
8. The method of claim 1 , wherein receiving the first segment of text data comprises receiving annotation data from an annotation file comprising at least one of a set of previously annotated model training data and a generated data model.
9. The method of claim 1 , further comprising generating at least one data model based on the at least one set of model training data and further based on at least one other set of previously annotated model training data selected according to a corresponding training descriptor.
10. The method of claim 1 , further comprising identifying weights assigned to the identified data features, and wherein predictively annotating the sequence of characters is performed based on the identified weights.
11. A system, comprising:
a processing unit;
a memory operatively coupled to the processing unit; and
a program module which executes in the processing unit from the memory and which, when executed by the processing unit, causes the system to perform machine learning management functions that include:
receiving a first segment of text data,
identifying data features corresponding to a sequence of characters in the first segment of text data,
generating predictive annotations to the sequence of characters based at least in part on the identified data features,
identifying inaccurate annotations generated according to the predictive annotations,
correcting the identified inaccurate annotations,
generating at least one set of model training data incorporating the corrected annotations, and
monitoring progress of annotations made to a second segment of text data associated with the first segment of text data by a plurality of collaborating users of a plurality of managed computers, the monitoring including determining, based at least in part on a training descriptor corresponding to the second segment of text data, a state of completion of annotations made to the second segment of text data by a particular one of the plurality of collaborating users, wherein the training descriptor identifies types of annotations in the at least one set of model training data.
12. The system of claim 11 , wherein at least one of the machine learning management functions is performed in response to receiving user input data that is received from a user via a user interface that is operatively coupled to the processing unit.
13. The system of claim 11 , wherein the state of completion of annotations made to the second segment of text data by the particular one of the plurality of collaborating users includes a state of incomplete, in-process annotation.
14. The system of claim 11 , wherein the training descriptor further identifies resource data corresponding to the at least one set of model training data, the resource data comprising a lexicon of associations between predefined sequences of characters and predefined labels.
15. The system of claim 11 , wherein the types of annotations identified by the training descriptor comprise at least one of categories and parts of speech associated with the sequence of characters.
16. The system of claim 11 , wherein the training descriptor further identifies types of features in the identified data features and a relative weight assigned to each of the types of features in the identified data features.
17. The system of claim 11 , wherein the machine learning management functions further include assigning the training descriptor to the generated set of model training data.
18. The system of claim 11 , wherein receiving the first segment of text data comprises receiving annotation data from an annotation file comprising at least one of a set of previously annotated model training data and a generated data model.
19. The system of claim 11 , wherein the machine learning management functions further include generating at least one data model based on the at least one set of model training data and further based on at least one other set of previously annotated model training data selected according to a corresponding training descriptor.
20. The system of claim 11 , wherein the machine learning management functions further include identifying weights assigned to the identified data features, and wherein predictively annotating the sequence of characters is performed based on the identified weights.
21. A non-transitory computer-readable storage medium having computer-executable instructions stored thereon which, when executed by a processing unit, cause a computer to perform machine learning management functions that include:
receiving a first segment of text data,
identifying data features corresponding to a sequence of characters in the first segment of text data,
generating predictive annotations to the sequence of characters based at least in part on the identified data features,
identifying inaccurate annotations generated according to the predictive annotations,
correcting the identified inaccurate annotations,
generating at least one set of model training data incorporating the corrected annotations, and
monitoring progress of annotations made to a second segment of text data associated with the first segment of text data by a plurality of collaborating users of a plurality of managed computers, the monitoring including determining, based at least in part on a training descriptor corresponding to the second segment of text data, a state of completion of annotations made to the second segment of text data by a particular one of the plurality of collaborating users, wherein the training descriptor identifies types of annotations in the at least one set of model training data.
22. The computer-readable storage medium of claim 21 , wherein at least one of the machine learning management functions is performed in response to receiving user input data that is received from a user via a user interface that is operatively coupled to the processing unit.
23. The computer-readable storage medium of claim 21 , wherein the state of completion of annotations made to the second segment of text data by the particular one of the plurality of collaborating users includes a state of incomplete, in-process annotation.
24. The computer-readable storage medium of claim 21 , wherein the training descriptor further identifies resource data corresponding to the at least one set of model training data, the resource data comprising a lexicon of associations between predefined sequences of characters and predefined labels.
25. The computer-readable storage medium of claim 21 , wherein the types of annotations identified by the training descriptor comprise at least one of categories and parts of speech associated with the sequence of characters.
26. The computer-readable storage medium of claim 21 , wherein the training descriptor further identifies types of features in the identified data features and a relative weight assigned to each of the types of features in the identified data features.
27. The computer-readable storage medium of claim 21 , wherein the machine learning management functions further include assigning the training descriptor to the generated set of model training data.
28. The computer-readable storage medium of claim 21 , wherein receiving the first segment of text data comprises receiving annotation data from an annotation file comprising at least one of a set of previously annotated model training data and a generated data model.
29. The computer-readable storage medium of claim 21 , wherein the machine learning management functions further include generating at least one data model based on the at least one set of model training data and further based on at least one other set of previously annotated model training data selected according to a corresponding training descriptor.
30. The computer-readable storage medium of claim 21 , wherein the machine learning management functions further include identifying weights assigned to the identified data features, and wherein predictively annotating the sequence of characters is performed based on the identified weights.