Parallel object-oriented decision tree system

• US 7,007,035 B2
• Filed: 06/08/2001
• Issued: 02/28/2006
• Est. Priority Date: 06/08/2001
• Status: Expired due to Fees

• Current Assignee: Lawrence Livermore National Security LLC (Government of the United States of America)
• Original Assignee: Regents of the University of California (University of California)
• Inventors: Cantu-Paz, Erick, Kamath, Chandrika
• Primary Examiner(s): AMSBURY, WAYNE P
• Application Number:
  US09/877,570
• Publication Number:
  US 20030061228A1
• Time in Patent Office: 1,726 Days
• Field of Search: 707/102, 706/45, 712/13
• US Class Current: 1/1
• CPC Class Codes: G06F 18/24323 Tree-organised classifiers
• CPC Class Codes: G06F 7/24 Sorting, i.e. extracting data from one or more carriers, rearranging the data in numerical or other ordered sequence, and rerecording the sorted data on the original carrier or on a different carrier or set of carriers sorting methods in generalG06F7/36 takes precedence
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A data mining decision tree system that uncovers patterns, associations, anomalies, and other statistically significant structures in data by reading and displaying data files, extracting relevant features for each of the objects, and using a method of recognizing patterns among the objects based upon object features through a decision tree that reads the data, sorts the data if necessary, determines the best manner to split the data into subsets according to some criterion, and splits the data.

Claims

1. A decision tree based data mining system for processing data;

comprising;

a multiplicity of processors,

an object oriented pattern recognition algorithms module for pattern recognition, comprising;

a multiplicity of data files,

a decision tree system including

an object oriented module operatively connected to said processors and connected to said data files to read said data and partition said data files among said multiplicity of processors,

an object oriented module operatively connected to said processors to parallel sort said data using said multiplicity of processors, if sorting is necessary,

an object oriented module operatively connected to said processors to determine the best manner to split said data according to some criterion wherein said criterion is the twoing rule, and

an object oriented module operatively connected to said processors to split said data, and

a data mining system, having

a storage module, and

an object oriented linking module for linking said decision tree system and said storage module.

2. A decision tree based data mining system for processing data, comprising:

a multiplicity of processors,

an object oriented pattern recognition algorithms module for pattern recognition, comprising;

a multiplicity of data files,

a decision tree system including

a parallel object oriented module operatively connected to said processors and connected to said data files to read said data and partition said data files among said multiplicity of processors, said data containing data items with features,

a parallel object oriented module operatively connected to said processors to parallel sort said data using said multiplicity of processors, if sorting is necessary,

a parallel object oriented module operatively connected to said processors to determine the best manner to split said data into subsets according to some criterion wherein said criterion is the twoing rule,

a parallel object oriented module operatively connected to said processors to split said data, and

a data mining system, having

a storage module to store the features for each data item,

a parallel object oriented linking module for linking said decision tree system and said storage module.

3. A decision tree based data mining method for processing data utilizing a multiplicity of processors, comprising the steps of:

providing data files containing objects having relevant features,

recognizing patterns among said objects based upon said relevant features,

creating a decision tree system,

using said multiplicity of processors for reading said data from said data files using an object oriented module,

using said multiplicity of processors for partitioning said data files among said multiplicity of processors,

using said multiplicity of processors for parallel sorting said data using an object oriented module and said multiplicity of processors if sorting is necessary,

determining the best manner to split said data into subsets according to some criterion using an object oriented module wherein said criterion is the twoing rule, and

splitting said data using an object oriented module.

4. A decision tree based data mining method for processing data utilizing a multiplicity of processors, comprising the steps of:

using said multiplicity of processor for reading and displaying data files, said data files containing objects having at least one feature,

partitioning said data files among said multiplicity of processors,

identifying said objects in said data files,

extracting at least one feature for each of said objects

recognizing patterns among said objects based upon said features, and

creating a decision tree, said decision tree including

using said multiplicity of processor for reading said data,

using said multiplicity of processor for parallel sorting said data using said multiplicity of processors, if sorting is necessary,

determining the best manner to split said data into subsets according to some criterion wherein said criterion is the twoing rule, and

splitting said data.