Method for constructing segmentation-based predictive models

• US 8,392,153 B2
• Filed: 09/04/2008
• Issued: 03/05/2013
• Est. Priority Date: 03/11/2002
• Status: Expired due to Fees

• Current Assignee: GlobalFoundries, Inc.
• Original Assignee: International Business Machines Corporation
• Inventors: Pednault, Edwin Peter Dawson, Natarajan, Ramesh
• Primary Examiner(s): Thangavelu, Kandasamy
• Assistant Examiner(s): OSBORNE, LUKE R
• Application Number:
  US12/204,670
• Publication Number:
  US 20090030864A1
• Time in Patent Office: 1,643 Days
• Field of Search: 703/2, 706/12, 706/15, 706/45, 707/101, 707/103.R
• US Class Current: 703/2
• CPC Class Codes: G06F 16/2465 Query processing support for facilitating data mining operations in structured databases
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06F 18/231 Hierarchical techniques, i.e. dividing or merging pattern sets so as to obtain a dendrogram
• CPC Class Codes: G06F 18/24323 Tree-organised classifiers
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/01 Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06V 10/7625 Hierarchical techniques, i.e. dividing or merging patterns to obtain a tree-like representation; Dendograms
• CPC Class Codes: G06V 10/764 using classification, e.g. of video objects
• CPC Class Codes: G06V 10/774 Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

The present invention generally relates to computer databases and, more particularly, to data mining and knowledge discovery. The invention specifically relates to a method for constructing segmentation-based predictive models, such as decision-tree classifiers, wherein data records are partitioned into a plurality of segments and separate predictive models are constructed for each segment. The present invention contemplates a computerized method for automatically building segmentation-based predictive models that substantially improves upon the modeling capabilities of decision trees and related technologies, and that automatically produces models that are competitive with, if not better than, those produced by data analysts and applied statisticians using traditional, labor-intensive statistical techniques. The invention achieves these properties by performing segmentation and multivariate statistical modeling within each segment simultaneously. Segments are constructed so as to maximize the accuracies of the predictive models within each segment. Simultaneously, the multivariate statistical models within each segment are refined so as to maximize their respective predictive accuracies.

Claims

1. A method for a process performed on a computer for training multivariate segment model objects, the method comprising:

1) accessing a collection of training data records comprising examples of input values that are available to a multivariate segment model object, together with corresponding desired output value(s) that a multivariate segment model is intended to predict;

2) presenting, as executed by a processor on a computer, the training data records to the multivariate segment model object by calling one or more scan-data-record interface functions, wherein the multivariate segment model object responds by generating and pruning pluralities of data segments and associated segment models, at least one of which comprises a training-data-based multivariate segment model for the multivariant segment model; and

3) repeating said accessing and said presenting until the multivariate segment model object indicates that it does not need to have the training records presented over again, as determined by calculating whether a predictive accuracy of the data segments and associated segment models is optimal, based on an evaluation using validation data records different from said training data records.

2. A non-transitory, signal-bearing storage medium tangibly embodying a program of machine-readable instructions executable by a digital processing apparatus to perform the method of claim 1.

3. A digital processing apparatus configured with a program of machine-readable instructions to perform the method of claim 1.

4. The method of claim 1, wherein said generating said plurality of data segments comprises at least one of:

a top-down decision-tree-like segmentation;

a bottom-up agglomerative-clustering-like segmentation;

a hybrid of said top-down decision-tree-like segmentation and said bottom-up agglomerative-clustering-like segmentation; and

a nearest-neighbor approach wherein a new data segment is defined based on a shortest distance between sets of seed records.

5. The method of claim 4, wherein said generating said plurality of data segments comprises at least one constraint placed on how said data segments are generated.

6. The method of claim 1, wherein an object-oriented embodiment is used for a module executing at least one of:

said generating said data segments;

said generating said segment model; and

said pruning.

7. The method of claim 1, wherein said generating of the plurality of associated segments models comprises:

varying degrees of freedom of a segment model.

8. The method of claim 7, wherein optimum degrees of freedom are determined using said training data records alone and holdout validation data is used for making a globally optimal choice among alternative segmentations and segment models.

9. The method of claim 1, wherein said pruning so as to optimize the predictive accuracy of a resulting segmentation-based predictive model comprises:

simultaneously considering both a number of generated data segments and degrees of freedom of the generated segment models.

10. The method of claim 1, wherein said pruning comprises:

selecting a subset of tree nodes and alternative models for each selected node so as to minimize a sum of degrees of fit.

11. The method of claim 10, wherein said degrees of fit comprise statistical confidence intervals for true expected values of said degrees of fit and said selecting further comprises consideration of upper confidence bounds for said true expected values.

12. The method of claim 10, wherein said selecting of alternative segment models is based upon only said training data, said selecting of alternative segment models comprising one of:

a cross validation of said training data; and

Bayesian scores as a measure of degree of fit.

13. The method of claim 1, wherein said one or more scan-data-record interface functions comprise one or more of:

an interface function for setting data scan options that determine how a model object is to be trained;

an interface function for informing a model object that a data scan is about to begin and for obtaining from the model object an indication of whether a data scan is needed by the model object;

an interface function for presenting data records to the model object; and

an interface function for informing the model object that a data scan is complete and for obtaining from the model object an indication of whether another data scan is needed by the model object.

14. An apparatus, comprising:

at least one processor; and

a memory interface to a database containing a collection of training data records comprising examples of input values that are available to a multivariate segment model object, together with corresponding desired output value(s) that a multivariate segment model is intended to predict, along with validation data records different from said training data records,

wherein said at least one processor executes a method for training multivariate segment model objects, the method comprising;

1) accessing the collection of training data records;

2) presenting the training data records to the multivariate segment model object by calling one or more scan-data-record interface functions, wherein the multivariate segment model object responds by generating and pruning pluralities of data segments and associated segment models, at least one of which comprises a training-data-based multivariate segment model for the multivariate segment model; and

3) repeating said accessing and said presenting until the multivariate segment model object indicates that it does not need to have the training records presented over again, as determined by calculating whether a predictive accuracy of the data segments and associated segment models is optimal, based on an evaluation using validation data records different from said training data records.

15. The apparatus of claim 14, wherein said one or more scan-data-record interface functions comprise one or more of:

an interface function for setting data scan options that determine how a model object is to be trained;

an interface function for informing a model object that a data scan is about to begin and for obtaining from the model object an indication of whether a data scan is needed by the model object;

an interface function for presenting data records to the model object; and

an interface function for informing the model object that a data scan is complete and for obtaining from the model object an indication of whether another data scan is needed by the model object.