Systems and techniques for predictive data analytics

• US 9,652,714 B2
• Filed: 07/22/2016
• Issued: 05/16/2017
• Est. Priority Date: 05/23/2014
• Status: Active Grant

• Current Assignee: DataRobot, Inc.
• Original Assignee: DataRobot, Inc.
• Inventors: Achin, Jeremy, DeGodoy, Thomas, Owen, Timothy, Conort, Xavier
• Primary Examiner(s): SITIRICHE, LUIS A
• Application Number:
  US15/217,626
• Publication Number:
  US 20160335550A1
• Time in Patent Office: 298 Days
• Field of Search: None
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/067 Enterprise or organisation modelling

Abstract

Systems and techniques for predictive data analytics are described. In a method for selecting a predictive model for a prediction problem, the suitabilities of predictive modeling procedures for the prediction problem may be determined based on characteristics of the prediction problem and/or on attributes of the respective modeling procedures. A subset of the predictive modeling procedures may be selected based on the determined suitabilities of the selected modeling procedures for the prediction problem. A resource allocation schedule allocating computational resources for execution of the selected modeling procedures may be generated, based on the determined suitabilities of the selected modeling procedures for the prediction problem. Results of the execution of the selected modeling procedures in accordance with the resource allocation schedule may be obtained. A predictive model for the prediction problem may be selected based on those results.

Claims

1. A predictive modeling apparatus comprising:

a memory configured to store a machine-executable module encoding a predictive modeling procedure, wherein the predictive modeling procedure includes a plurality of tasks, wherein the machine-executable module includes a directed graph representing dependencies between the tasks, and wherein the plurality of tasks includes at least one pre-processing task, at least one model-fitting task, and at least one post-processing task; and

at least one processor configured to execute the machine-executable module, wherein executing the machine-executable module causes the apparatus to perform the predictive modeling procedure, including;

manipulating input data, comprising performing the pre-processing task on the input data;

performing the model-fitting task, comprising;

generating, from the pre-processed input data, training data and testing data,

fitting a predictive model to the training data, and

testing the fitted model on the testing data; and

performing the post-processing task,

wherein the pre-processed input data comprise at least one data set, wherein generating the training data comprises obtaining a first subset of the data set, and wherein generating the testing data comprises obtaining a second subset of the data set,

wherein performing the predictive modeling procedure further includes performing cross-validation of the predictive model,

wherein the training data are first training data, wherein the testing data are first testing data, wherein the fitted model is a first fitted model, and wherein performing the cross-validation of the predictive model comprises;

(a) generating, from the data set, second training data and second testing data, wherein generating the second training data comprises obtaining a third subset of the data set, and wherein generating the second testing data comprises obtaining a fourth subset of the data set;

(b) fitting the predictive model to the second training data to obtain a second fitted model; and

(c) testing the second fitted model on the second testing data.

2. The predictive modeling apparatus of claim 1, wherein performing the model-fitting task further comprises partitioning the data set into a plurality of partitions including at least a first partition and a second partition.

3. The predictive modeling apparatus of claim 2, wherein partitioning the data set into a plurality of partitions comprises randomly assigning each data item in the data set to a respective partition.

4. The predictive modeling apparatus of claim 2, wherein:

the first training data comprise the first partition of the data set;

the first testing data comprise all of the partitions of the data set except the first partition;

the second training data comprise the second partition of the data set; and

the second testing data comprise all of the partitions of the data set except the second partition.

5. The predictive modeling apparatus of claim 2, wherein:

the first training data comprise a subset of the first partition of the data set;

the first testing data comprise respective subsets of all of the partitions of the data set except the first partition;

the second training data comprise a subset of the second partition of the data set; and

the second testing data comprise respective subsets of all of the partitions of the data set except the second partition.

6. The predictive modeling apparatus of claim 1, wherein:

the pre-processed input data comprise a first partition and a second partition,

the data set comprises the first partition of the pre-processed input data, and

performing the model-fitting task further comprises testing the first and second fitted models on holdout data comprising the second partition of the pre-processed input data.

7. The predictive modeling apparatus of claim 6, wherein no predictive model is fitted to the holdout data.

8. The predictive modeling apparatus of claim 1, wherein the at least one processor is further configured to refit the predictive model and deploy the refitted model.

9. The predictive modeling apparatus of claim 8, wherein deploying the refitted model comprises generating a plurality of predictions by applying the refitted model to other data representing instances of a prediction problem, wherein the input data do not include the other data.

10. The predictive modeling apparatus of claim 8, wherein deploying the refitted model comprises caching a state of the refitted model in a memory accessible to the at least one processor.

11. The predictive modeling apparatus of claim 8, wherein the refitted model has a first representation, and wherein deploying the refitted model comprises:

generating a second representation of the refitted model, wherein the second representation comprises a set of one or more conditional rules,

wherein the second representation of the refitted model is a machine executable representation, and wherein the set of one or more conditional rules comprises a set of one or more machine executable if-then statements.

12. The predictive modeling apparatus of claim 8, wherein the input data are first input data, and wherein deploying the refitted model further comprises refreshing the refitted model based, at least in part, on second input data.

13. The predictive modeling apparatus of claim 12, wherein the refitted model is a first refitted model, and wherein refreshing the refitted model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on the second input data to generate a second refitted model; and

blending the first refitted model and the second refitted model to generate a refreshed predictive model.

14. The predictive modeling apparatus of claim 12, wherein refreshing the refitted model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on third input data comprising at least a portion of the first input data and at least a portion of the second input data to generate a refreshed predictive model.

15. The predictive modeling apparatus of claim 1, wherein the predictive model is a first type of predictive model, wherein the model-fitting task is a first model-fitting task, and wherein performing the predictive modeling procedure further includes:

performing a second model-fitting task using a second type of predictive model, wherein performing the second model-fitting task includes fitting the second type of predictive model to obtain a second fitted model; and

blending the first fitted model of the first type and the second fitted model of the second type to generate a blended model.

16. A predictive modeling apparatus comprising:

a memory configured to store a machine-executable module encoding a predictive modeling procedure, wherein the predictive modeling procedure includes a plurality of tasks, wherein the machine-executable module includes a directed graph representing dependencies between the tasks, and wherein the plurality of tasks includes at least one pre-processing task, at least one model-fitting task, and at least one post-processing task; and

at least one processor configured to execute the machine-executable module, wherein executing the machine-executable module causes the apparatus to perform the predictive modeling procedure, including;

manipulating input data, comprising performing the pre-processing task on the input data;

performing the model-fitting task, comprising;

generating, from the pre-processed input data, training data and testing data,

fitting a predictive model to the training data, and

testing the fitted model on the testing data; and

performing the post-processing task,

wherein performing the predictive modeling procedure further includes performing nested cross-validation of the predictive model, and wherein;

the pre-processed input data comprise at least one data set;

performing the nested cross-validation of the predictive model comprises;

partitioning the data set into a first plurality of partitions of the data set including at least a first partition of the data set and a second partition of the data set, and

partitioning the first partition of the data set into a plurality of partitions of the first partition of the data set including at least a first partition of the first partition of the data set and a second partition of the first partition of the data set;

the training data comprise the first partition of the first partition of the data set; and

the testing data comprise all of the partitions of the first partition of the data set except the first partition of the first partition of the data set.

17. The predictive modeling apparatus of claim 16, wherein the training data are first training data, the testing data are first testing data, the fitted model is a first fitted model, and performing the nested cross-validation of the predictive model further comprises:

(a) generating, from the first partition of the data set, second training data and second testing data, wherein the second training data comprise the second partition of the first partition of the data set, and wherein the second testing data comprise a plurality of the partitions of the first partition of the data set other than the second partition of the first partition of the data set;

(b) fitting the predictive model to the second training data to obtain a second fitted model; and

(c) testing the second fitted model on the second testing data.

18. The predictive modeling apparatus of claim 17, wherein performing the nested cross-validation further includes:

testing the first fitted model and the second fitted model on the second partition of the data set; and

comparing the first fitted model to the second fitted model based on results of testing the first and second fitted models on the second partition of the data set.

19. The predictive modeling apparatus of claim 16, wherein the at least one processor is further configured to refit the predictive model and deploy the refitted model.

20. The predictive modeling apparatus of claim 19, wherein deploying the refitted model comprises generating a plurality of predictions by applying the refitted model to other data representing instances of a prediction problem, wherein the input data do not include the other data.

21. The predictive modeling apparatus of claim 19, wherein deploying the refitted model comprises caching a state of the refitted model in a memory accessible to the at least one processor.

22. The predictive modeling apparatus of claim 19, wherein the refitted model has a first representation, and wherein deploying the refitted model comprises:

generating a second representation of the refitted model, wherein the second representation comprises a set of one or more conditional rules,

wherein the second representation of the refitted model is a machine executable representation, and wherein the set of one or more conditional rules comprises a set of one or more machine executable if-then statements.

23. The predictive modeling apparatus of claim 19, wherein the input data are first input data, and wherein deploying the refitted model further comprises refreshing the refitted model based, at least in part, on second input data.

24. The predictive modeling apparatus of claim 23, wherein the refitted model is a first refitted model, and wherein refreshing the refitted model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on the second input data to generate a second refitted model; and

blending the first refitted model and the second refitted model to generate a refreshed predictive model.

25. The predictive modeling apparatus of claim 23, wherein refreshing the refitted model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on third input data comprising at least a portion of the first input data and at least a portion of the second input data to generate a refreshed predictive model.

26. The predictive modeling apparatus of claim 16, wherein the predictive model is a first type of predictive model, wherein the fitted model is a first fitted model, wherein the model-fitting task is a first model-fitting task, and wherein performing the predictive modeling procedure further includes:

performing a second model-fitting task using a second type of predictive model, wherein performing the second model-fitting task comprises fitting the second type of predictive model to obtain a second fitted model; and

blending the first fitted model of the first type and the second fitted model of the second type to generate a blended model.

27. A predictive modeling apparatus comprising:

a memory configured to store a machine-executable module encoding a predictive modeling procedure, wherein the predictive modeling procedure includes a plurality of tasks, wherein the machine-executable module includes a directed graph representing dependencies between the tasks, and wherein the plurality of tasks includes at least one pre-processing task, at least one model-fitting task, and at least one post-processing task; and

at least one processor configured to execute the machine-executable module, wherein executing the machine-executable module causes the apparatus to perform the predictive modeling procedure, including;

manipulating input data, comprising performing the pre-processing task on the input data performing the model-fitting task, comprising;

generating, from the pre-processed input data, training data and testing data,

fitting a predictive model to the training data, and

testing the fitted model on the testing data; and

performing the post-processing task,

wherein the pre-processed input data comprise at least one data set, wherein generating the training data comprises obtaining a first subset of the data set, and wherein generating the testing data comprises obtaining a second subset of the data set,

wherein the predictive model is a first type of predictive model, the fitted model is a first fitted model, the model-fitting task is a first model-fitting task, and performing the predictive modeling procedure further includes performing a second model-fitting task using a second type of predictive model.

28. The predictive modeling apparatus of claim 27, wherein the training data are first training data, wherein the testing data are first testing data, and wherein performing the second model-fitting task using the second type of predictive model comprises:

(a) generating, from the data set, second training data and second testing data, wherein generating the second training data comprises obtaining a third subset of the data set, and wherein generating the second testing data comprises generating a fourth subset of the data set;

(b) fitting the second type of predictive model to the second training data to obtain a second fitted model; and

(c) testing the second fitted model on the second testing data.

29. The predictive modeling apparatus of claim 28, wherein performing the predictive modeling procedure further includes blending the first fitted model of the first type and the second fitted model of the second type to generate a blended model.

30. The predictive modeling apparatus of claim 29, wherein:

the pre-processed input data comprise a first partition and a second partition,

the data set comprises the first partition of the pre-processed input data, and

blending the first and second fitted models comprises;

fitting the blended model to a first partition of the second partition of the pre-processed input data, and

testing the fitted blended model on a second partition of the second partition of the pre-processed input data.

31. The predictive modeling apparatus of claim 29, wherein the at least one processor is further configured to deploy the blended model.

32. The predictive modeling apparatus of claim 31, wherein deploying the blended model comprises generating a plurality of predictions by applying the blended model to other data representing instances of a prediction problem, wherein the input data do not include the other data.

33. The predictive modeling apparatus of claim 31, wherein deploying the blended model comprises caching a state of the blended model in a memory accessible to the at least one processor.

34. The predictive modeling apparatus of claim 31, wherein the blended model has a first representation, and wherein deploying the blended model comprises:

generating a second representation of the blended model, wherein the second representation comprises a set of one or more conditional rules,

wherein the second representation of the blended model is a machine executable representation, and wherein the set of one or more conditional rules comprises a set of one or more machine executable if-then statements.

35. The predictive modeling apparatus of claim 31, wherein the input data are first input data, and wherein deploying the blended model further comprises refreshing the blended model based, at least in part, on second input data.

36. The predictive modeling apparatus of claim 35, wherein the blended model is a first blended model, and wherein refreshing the blended model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on the second input data to generate a second blended model; and

blending the first blended model and the second blended model to generate a refreshed predictive model.

37. The predictive modeling apparatus of claim 36, wherein refreshing the blended model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on third input data comprising at least a portion of the first input data and at least a portion of the second input data to generate a refreshed blended model.

38. A predictive modeling apparatus comprising:

a memory configured to store a machine-executable module encoding a predictive modeling procedure, wherein the predictive modeling procedure includes a plurality of tasks, wherein the machine-executable module includes a directed graph representing dependencies between the tasks, and wherein the plurality of tasks includes at least one pre-processing task, at least one model-fitting task, and at least one post-processing task; and

at least one processor configured to execute the machine-executable module, wherein executing the machine-executable module causes the apparatus to perform the predictive modeling procedure, including;

manipulating input data, comprising performing the pre-processing task on the input data;

performing the model-fitting task, comprising;

generating, from the pre-processed input data, training data and testing data,

fitting a predictive model to the training data, and

testing the fitted model on the testing data; and

performing the post-processing task,

wherein the at least one processor is further configured to deploy the fitted model,

wherein the fitted model has a first representation, and wherein deploying the fitted model comprises;

generating a second representation of the fitted model, wherein the second representation comprises a set of one or more conditional rules,

wherein the second representation of the fitted model is a machine executable representation, and wherein the set of one or more conditional rules comprises a set of one or more machine executable if-then statements.

39. The predictive modeling apparatus of claim 38, wherein deploying the fitted model further comprises generating a plurality of predictions by applying the fitted model to other data representing instances of a prediction problem, wherein the input data do not include the other data.

40. The predictive modeling apparatus of claim 38, wherein deploying the fitted model further comprises caching a state of the fitted model in a memory accessible to the at least one processor.

41. A predictive modeling apparatus comprising:

a memory configured to store a machine-executable module encoding a predictive modeling procedure, wherein the predictive modeling procedure includes a plurality of tasks, wherein the machine-executable module includes a directed graph representing dependencies between the tasks, and wherein the plurality of tasks includes at least one pre-processing task, at least one model-fitting task, and at least one post-processing task; and

at least one processor configured to execute the machine-executable module, wherein executing the machine-executable module causes the apparatus to perform the predictive modeling procedure, including;

manipulating input data, comprising performing the pre-processing task on the input data;

performing the model-fitting task, comprising;

generating, from the pre-processed input data, training data and testing data,

fitting a predictive model to the training data, and

testing the fitted model on the testing data; and

performing the post-processing task,

wherein the at least one processor is further configured to deploy the fitted model,

wherein the input data are first input data, and wherein deploying the fitted model further comprises refreshing the fitted model based, at least in part, on second input data.

42. The predictive modeling apparatus of claim 41, wherein the fitted model is a first fitted model, and wherein refreshing the fitted model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on the second input data to generate a second fitted model; and

blending the first fitted model and the second fitted model to generate a refreshed predictive model.

43. The predictive modeling apparatus of claim 41, wherein refreshing the fitted model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on third input data comprising at least a portion of the first input data and at least a portion of the second input data to generate a refreshed predictive model.

44. The predictive modeling apparatus of claim 41, wherein performing the pre-processing task on the input data comprises cleansing the input data, performing feature selection on the input data, and/or performing feature engineering on the input data.

45. The predictive modeling apparatus of claim 41, wherein performing the model-fitting task further comprises selecting the predictive model from a plurality of predictive models, determining a value of at least one parameter of the predictive model, and/or determining a value of at least one hyper-parameter of the predictive modeling procedure.

46. The predictive modeling apparatus of claim 41, wherein performing the post-processing task comprises processing the tested model.

47. The predictive modeling apparatus of claim 46, wherein processing the tested model comprises calibrating the tested model and/or blending the tested model with another predictive model to generate a blended predictive model.

48. A predictive modeling method comprising:

accessing a machine-executable module encoding a predictive modeling procedure, wherein the predictive modeling procedure includes a plurality of tasks, wherein the machine-executable module includes a directed graph representing dependencies between the tasks, and wherein the plurality of tasks includes at least one pre-processing task, at least one model-fitting task, and at least one post-processing task; and

executing the machine-executable module, wherein executing the machine-executable module comprises performing the predictive modeling procedure, including;

manipulating input data, comprising performing the pre-processing task on the input data;

performing the model-fitting task, comprising;

generating, from the pre-processed input data, training data and testing data,

fitting a predictive model to the training data, and

testing the fitted model on the testing data; and

performing the post-processing task,

wherein the pre-processed input data comprise at least one data set, wherein generating the training data comprises obtaining a first subset of the data set, and wherein generating the testing data comprises obtaining a second subset of the data set,

wherein performing the predictive modeling procedure further includes performing cross-validation of the predictive model,

wherein the training data are first training data, wherein the testing data are first testing data, wherein the fitted model is a first fitted model, and wherein performing the cross-validation of the predictive model comprises;

(a) generating, from the data set, second training data and second testing data, wherein generating the second training data comprises obtaining a third subset of the data set, and wherein generating the second testing data comprises obtaining a fourth subset of the data set;

(b) fitting the predictive model to the second training data to obtain a second fitted model; and

(c) testing the second fitted model on the second testing data.

49. The method of claim 48, wherein performing the model-fitting task further comprises partitioning the data set into a plurality of partitions including at least a first partition and a second partition.

50. The method of claim 49, wherein partitioning the data set into a plurality of partitions comprises randomly assigning each data item in the data set to a respective partition.

51. The method of claim 49, wherein:

the first training data comprise the first partition of the data set;

the first testing data comprise all of the partitions of the data set except the first partition;

the second training data comprise the second partition of the data set; and

the second testing data comprise all of the partitions of the data set except the second partition.

52. The method of claim 49, wherein:

the first training data comprise a subset of the first partition of the data set;

the first testing data comprise respective subsets of all of the partitions of the data set except the first partition;

the second training data comprise a subset of the second partition of the data set; and

the second testing data comprise respective subsets of all of the partitions of the data set except the second partition.

53. The method of claim 48, wherein:

the pre-processed input data comprise a first partition and a second partition,

the data set comprises the first partition of the pre-processed input data, and

performing the model-fitting task further comprises testing the first and second fitted models on holdout data comprising the second partition of the pre-processed input data.

54. The method of claim 53, wherein no predictive model is fitted to the holdout data.

55. The method of claim 48, further comprising refitting the predictive model and deploying the refitted model.

56. The method of claim 55, wherein deploying the refitted model comprises generating a plurality of predictions by applying the refitted model to other data representing instances of a prediction problem, wherein the input data do not include the other data.

57. The method of claim 55, wherein deploying the refitted model comprises caching a state of the refitted model in a memory accessible to the at least one processor.

58. The method of claim 55, wherein the refitted model has a first representation, and wherein deploying the refitted model comprises:

generating a second representation of the refitted model, wherein the second representation comprises a set of one or more conditional rules,

wherein the second representation of the refitted model is a machine executable representation, and wherein the set of one or more conditional rules comprises a set of one or more machine executable if-then statements.

59. The method of claim 55, wherein the input data are first input data, and wherein deploying the refitted model further comprises refreshing the refitted model based, at least in part, on second input data.

60. The method of claim 59, wherein the refitted model is a first refitted model, and wherein refreshing the refitted model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on the second input data to generate a second refitted model; and

blending the first refitted model and the second refitted model to generate a refreshed predictive model.

61. The method of claim 59, wherein refreshing the refitted model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on third input data comprising at least a portion of the first input data and at least a portion of the second input data to generate a refreshed predictive model.

62. The method of claim 48, wherein the predictive model is a first type of predictive model, wherein the model-fitting task is a first model-fitting task, and wherein performing the predictive modeling procedure further includes:

performing a second model-fitting task using a second type of predictive model, wherein performing the second model-fitting task includes fitting the second type of predictive model to obtain a second fitted model; and

blending the first fitted model of the first type and the second fitted model of the second type to generate a blended model.

63. A predictive modeling method comprising:

accessing a machine-executable module encoding a predictive modeling procedure, wherein the predictive modeling procedure includes a plurality of tasks, wherein the machine-executable module includes a directed graph representing dependencies between the tasks, and wherein the plurality of tasks includes at least one pre-processing task, at least one model-fitting task, and at least one post-processing task; and

executing the machine-executable module, wherein executing the machine-executable module comprises performing the predictive modeling procedure, including;

manipulating input data, comprising performing the pre-processing task on the input data;

performing the model-fitting task, comprising;

generating, from the pre-processed input data, training data and testing data,

fitting a predictive model to the training data, and

testing the fitted model on the testing data; and

performing the post-processing task,

wherein performing the predictive modeling procedure further includes performing nested cross-validation of the predictive model, and wherein;

the pre-processed input data comprise at least one data set;

performing the nested cross-validation of the predictive model comprises;

partitioning the data set into a first plurality of partitions of the data set including at least a first partition of the data set and a second partition of the data set, and

partitioning the first partition of the data set into a plurality of partitions of the first partition of the data set including at least a first partition of the first partition of the data set and a second partition of the first partition of the data set;

the training data comprise the first partition of the first partition of the data set; and

the testing data comprise all of the partitions of the first partition of the data set except the first partition of the first partition of the data set.

64. The method of claim 63, wherein the training data are first training data, the testing data are first testing data, the fitted model is a first fitted model, and performing the nested cross-validation of the predictive model further comprises:

(a) generating, from the first partition of the data set, second training data and second testing data, wherein the second training data comprise the second partition of the first partition of the data set, and wherein the second testing data comprise a plurality of the partitions of the first partition of the data set other than the second partition of the first partition of the data set;

(b) fitting the predictive model to the second training data to obtain a second fitted model; and

(c) testing the second fitted model on the second testing data.

65. The method of claim 64, wherein performing the nested cross-validation further includes:

testing the first fitted model and the second fitted model on the second partition of the data set; and

comparing the first fitted model to the second fitted model based on results of testing the first and second fitted models on the second partition of the data set.

66. The method of claim 63, further comprising refitting the predictive model and deploying the refitted model.

67. The method of claim 66, wherein deploying the refitted model comprises generating a plurality of predictions by applying the refitted model to other data representing instances of a prediction problem, wherein the input data do not include the other data.

68. The method of claim 66, wherein deploying the refitted model comprises caching a state of the refitted model in a memory accessible to the at least one processor.

69. The method of claim 66, wherein the refitted model has a first representation, and wherein deploying the refitted model comprises:

generating a second representation of the refitted model, wherein the second representation comprises a set of one or more conditional rules,

wherein the second representation of the refitted model is a machine executable representation, and wherein the set of one or more conditional rules comprises a set of one or more machine executable if-then statements.

70. The method of claim 66, wherein the input data are first input data, and wherein deploying the refitted model further comprises refreshing the refitted model based, at least in part, on second input data.

71. The method of claim 70, wherein the refitted model is a first refitted model, and wherein refreshing the refitted model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on the second input data to generate a second refitted model; and

blending the first refitted model and the second refitted model to generate a refreshed predictive model.

72. The method of claim 70, wherein refreshing the refitted model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on third input data comprising at least a portion of the first input data and at least a portion of the second input data to generate a refreshed predictive model.

73. The method of claim 63, wherein the predictive model is a first type of predictive model, wherein the fitted model is a first fitted model, wherein the model-fitting task is a first model-fitting task, and wherein performing the predictive modeling procedure further includes:

performing a second model-fitting task using a second type of predictive model, wherein performing the second model-fitting task comprises fitting the second type of predictive model to obtain a second fitted model; and

blending the first fitted model of the first type and the second fitted model of the second type to generate a blended model.

74. A predictive modeling method comprising:

accessing a machine-executable module encoding a predictive modeling procedure, wherein the predictive modeling procedure includes a plurality of tasks, wherein the machine-executable module includes a directed graph representing dependencies between the tasks, and wherein the plurality of tasks includes at least one pre-processing task, at least one model-fitting task, and at least one post-processing task; and

executing the machine-executable module, wherein executing the machine-executable module comprises performing the predictive modeling procedure, including;

manipulating input data, comprising performing the pre-processing task on the input data;

performing the model-fitting task, comprising;

generating, from the pre-processed input data, training data and testing data,

fitting a predictive model to the training data, and

testing the fitted model on the testing data; and

performing the post-processing task,

wherein the pre-processed input data comprise at least one data set, wherein generating the training data comprises obtaining a first subset of the data set, and wherein generating the testing data comprises obtaining a second subset of the data set,

wherein the predictive model is a first type of predictive model, the fitted model is a first fitted model, the model-fitting task is a first model-fitting task, and performing the predictive modeling procedure further includes performing a second model-fitting task using a second type of predictive model.

75. The method of claim 74, wherein the training data are first training data, wherein the testing data are first testing data, and wherein performing the second model-fitting task using the second type of predictive model comprises:

(a) generating, from the data set, second training data and second testing data, wherein generating the second training data comprises obtaining a third subset of the data set, and wherein generating the second testing data comprises generating a fourth subset of the data set;

(b) fitting the second type of predictive model to the second training data to obtain a second fitted model; and

(c) testing the second fitted model on the second testing data.

76. The method of claim 75, wherein performing the predictive modeling procedure further includes blending the first fitted model of the first type and the second fitted model of the second type to generate a blended model.

77. The method of claim 76, wherein:

the pre-processed input data comprise a first partition and a second partition,

the data set comprises the first partition of the pre-processed input data, and

blending the first and second fitted models comprises;

fitting the blended model to a first partition of the second partition of the pre-processed input data, and

testing the fitted blended model on a second partition of the second partition of the pre-processed input data.

78. The method of claim 76, further comprising deploying the blended model.

79. The method of claim 78, wherein deploying the blended model comprises generating a plurality of predictions by applying the blended model to other data representing instances of a prediction problem, wherein the input data do not include the other data.

80. The method of claim 78, wherein deploying the blended model comprises caching a state of the blended model in a memory accessible to the at least one processor.

81. The method of claim 78, wherein the blended model has a first representation, and wherein deploying the blended model comprises:

generating a second representation of the blended model, wherein the second representation comprises a set of one or more conditional rules,

wherein the second representation of the blended model is a machine executable representation, and wherein the set of one or more conditional rules comprises a set of one or more machine executable if-then statements.

82. The method of claim 78, wherein the input data are first input data, and wherein deploying the blended model further comprises refreshing the blended model based, at least in part, on second input data.

83. The method of claim 82, wherein the blended model is a first blended model, and wherein refreshing the blended model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on the second input data to generate a second blended model; and

blending the first blended model and the second blended model to generate a refreshed predictive model.

84. The method of claim 82, wherein refreshing the blended model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on third input data comprising at least a portion of the first input data and at least a portion of the second input data to generate a refreshed blended model.

85. A predictive modeling method comprising:

accessing a machine-executable module encoding a predictive modeling procedure, wherein the predictive modeling procedure includes a plurality of tasks, wherein the machine-executable module includes a directed graph representing dependencies between the tasks, and wherein the plurality of tasks includes at least one pre-processing task, at least one model-fitting task, and at least one post-processing task;

executing the machine-executable module, wherein executing the machine-executable module comprises performing the predictive modeling procedure, including;

manipulating input data, comprising performing the pre-processing task on the input data;

performing the model-fitting task, comprising;

generating, from the pre-processed input data, training data and testing data,

fitting a predictive model to the training data, and

testing the fitted model on the testing data; and

performing the post-processing task; and

deploying the fitted model,

wherein the fitted model has a first representation, and wherein deploying the fitted model comprises;

generating a second representation of the fitted model, wherein the second representation comprises a set of one or more conditional rules, and

wherein the second representation of the fitted model is a machine executable representation, and wherein the set of one or more conditional rules comprises a set of one or more machine executable if-then statements.

86. The method of claim 85, wherein deploying the fitted model further comprises generating a plurality of predictions by applying the fitted model to other data representing instances of a prediction problem, wherein the input data do not include the other data.

87. The method of claim 85, wherein deploying the fitted model further comprises caching a state of the fitted model in a memory accessible to the at least one processor.

88. A predictive modeling method comprising:

accessing a machine-executable module encoding a predictive modeling procedure, wherein the predictive modeling procedure includes a plurality of tasks, wherein the machine-executable module includes a directed graph representing dependencies between the tasks, and wherein the plurality of tasks includes at least one pre-processing task, at least one model-fitting task, and at least one post-processing task;

executing the machine-executable module, wherein executing the machine-executable module comprises performing the predictive modeling procedure, including;

manipulating input data, comprising performing the pre-processing task on the input data;

performing the model-fitting task, comprising;

generating, from the pre-processed input data, training data and testing data,

fitting a predictive model to the training data, and

testing the fitted model on the testing data;

performing the post-processing task; and

deploying the fitted model,

wherein the input data are first input data, and wherein deploying the fitted model further comprises refreshing the fitted model based, at least in part, on second input data.

89. The method of claim 88, wherein the fitted model is a first fitted model, and wherein refreshing the fitted model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on the second input data to generate a second fitted model; and

blending the first fitted model and the second fitted model to generate a refreshed predictive model.

90. The method of claim 88, wherein refreshing the fitted model based, at least in part, on the second input data comprises:

performing the predictive modeling procedure on third input data comprising at least a portion of the first input data and at least a portion of the second input data to generate a refreshed predictive model.