Computerized data mining system and program product

Abstract

Under the present invention, a data exploration system, a customized model system and an existing model system are provided. The data exploration system analyzes user data to identify statistical information such as data distribution, data relationships, data outliners and invalid or missing data values. The customized model center iteratively generates customized data mining models in parallel based on permutations of the user data, user-provided business parameters and/or a set of model generation algorithms. The existing model system provides users with a library of existing data mining models, assembled based on the business parameters, from which they can choose one or more. In any event, any customized or existing data mining models selected can be run against the user data in parallel.

1 . A computerized data mining system, comprising:
 a central processing unit;   a memory operably associated with the central processing unit; and   a data mining system storable in the memory and executable by the central processing unit, the data mining system comprising:
 a data exploration system for receiving and analyzing user data to provide statistical information about the user data, wherein the statistical information comprises data relationships, data outliners, invalid data values and standard deviations; 
 a customized model system for generating and ranking customized data mining models, and for executing a selected customized data mining model on the user data, wherein the customized data mining models are generated using multiple iterations based on permutations of at least one of the user data, business parameters and a set of model generation algorithms, wherein the business parameters comprise a business taxonomy and a set of model goals, wherein the customized model system comprises:
 a model generation system for generating the customized data mining models in parallel using multiple iterations based on the permutations of at least one of the user data, the business parameters and the set of model generation algorithms; 
 a model ranking system for ranking the customized data mining models based on the business parameters, for identifying a predetermined quantity of the ranked customized data mining models, and for providing comparative data corresponding to the predetermined quantity of the ranked customized data mining models; 
 
 a customized model selection system for selecting at least one customized mining model from the predetermined quantity; and 
 a customized model execution system for executing the selected at least one customized data mining model on the user data; and 
 an existing model system for selecting at least one existing data mining model from a library of existing data mining models, and for executing the selected at least one existing data mining model in parallel on the user data, and outputting a result of the executing of the selected at least one customized data mining model to a user. 
   
   
   
       2 . The computerized data mining system of  claim 1 , further comprising:
 a data submission system for submitting the user data; and   a parameter designation system for designating the business parameters.   
   
   
       3 . The computerized data mining system of  claim 1 , wherein the existing model system comprises:
 a model library system for assembling the library of existing data mining models based on the business parameters, and for displaying the library of existing data mining models and comparative data corresponding to the library of existing data models;   
     an existing model selection system for selecting the at least one existing data mining model from the library of existing data mining models;
 an existing model execution system for executing the at least one existing data mining model on the user data in parallel; and 
 a existing model comparison system for comparing results of the execution of the at least one existing data mining model. 
 
   
   
       4 . The computerized data mining system of  claim 1 , wherein the business parameters comprise a business taxonomy, a set of business problems and a set of model goals. 
   
   
       5 . The computerized data mining system of  claim 1 , wherein the statistical information comprises data relationships, data outliners, invalid data values and standard deviations. 
   
   
       6 . The computerized data mining system of  claim 1 , wherein the computerized data mining system is implemented in a network environment. 
   
   
       7 . A computer-readable storage medium storing computer instructions, which when executed, enables a computer system to mine data, the computer instructions comprising:
 receiving and analyzing user data to provide statistical information about the user data, wherein the statistical information comprises data relationships, data outliners, invalid data values and standard deviations;   generating and ranking customized data mining models, and executing a selected customized data mining model on the user data, wherein the customized data mining models are generated using multiple iterations based on permutations of at least one of the user data, business parameters and a set of model generation algorithms, wherein the generating and ranking comprises:
 generating the customized data mining models in parallel using multiple iterations based on the permutations of at least one of the user data, the business parameters and the set of model generation algorithms, wherein the business parameters comprise a business taxonomy and a set of model goals; 
 ranking the customized data mining models based on the business parameters, identifying a predetermined quantity of the ranked customized data mining models, and providing comparative data corresponding to the predetermined quantity of the ranked customized data mining models; 
   selecting at least one customized mining model from the predetermined_quantity;   
     executing the selected at least one customized data mining model on the user data; and
 selecting at least one existing data mining model from a library of existing data mining models, and executing the selected at least one existing data mining model in parallel on the user data, and outputting a result of the executing of the selected at least one customized data mining model to a user. 
 
   
   
       8 . The computer-readable storage medium storing computer instructions of  claim 7 , wherein the instructions further comprise:
 submitting the user data; and   designating the business parameters.   
   
   
       9 . The computer-readable storage medium storing computer instructions of  claim 7 , wherein the selecting comprises:
 assembling the library of existing data mining models based on the business parameters, and for displaying the library of existing data mining models and comparative data corresponding to the library of existing data mining models;   selecting the at least one existing data mining model from the library of existing data mining models;   executing the at least one existing data mining model on the user data in parallel; and   comparing results of the execution of the at least one existing data mining model.   
   
   
       10 . The computer-readable storage medium storing computer instructions of  claim 7 , wherein the business parameters comprise a business taxonomy, a set of business problems and a set of model goals. 
   
   
       11 . The computer-readable storage medium storing computer instructions of  claim 7 , wherein the computer instructions are implemented in a network environment.