Method and system for interactive decision tree modification and visualization

1. A method in a server coupled to a client for modifying an original data mining model for display at the client, the original data mining model maintained by the server, the method comprising:

receiving a request message at the server from the client to modify the original data mining model for use at the client, wherein the request message is generated at the client and includes at least one command for modifying the original data mining model for use at the client;

generating change definitions at the server from the request message to modify the original data mining model for use at the client, wherein the change definitions define only changes to the original data mining model resulting from application of the at least one command to the original data mining model at the server;

storing the change definitions in a document at the server; and,

transmitting the document from the server to the client, wherein the client generates a modified data mining model to replace the original data mining model at the client and for display at the client using the change definitions received from the server, whereby application of the at least one command to the original data mining model is performed at the server rather than at the client and only changes to the original data mining model resulting from the application of the at least one command to the original data mining model at the server are transmitted from the server to the client;

wherein the document is a mark-up language document; and,

wherein each of the original data mining model and the modified data mining model is a decision tree.

2. The method of claim 1 wherein the request message is a syntactically extended structured query language command.

3. The method of claim 1 wherein the request message includes at least one of:

an erase split command to remove a portion of the decision tree;

a find split command to grow a portion of the decision tree;

a goto split command to view an alternative decision tree; and,

a force split command to shape the decision tree in accordance with user specifications.

4. A method in a server coupled to a client for modifying an original data mining model for display at the client, the original data mining model maintained by the server, the method comprising:

receiving a request message at the server from the client to modify the original data mining model for use at the client, wherein the request message is generated at the client and includes at least one command for modifying the original data mining model for use at the client;

generating change definitions at the server from the request message to modify the original data mining model for use at the client, wherein the change definitions define only changes to the original data mining model resulting from application of the at least one command to the original data mining model at the server;

storing the change definitions in a document at the server; and,

transmitting the document from the server to the client, wherein the client generates a modified data mining model to replace the original data mining model at the client and for display at the client using the change definitions received from the server, whereby application of the at least one command to the original data mining model is performed at the server rather than at the client and only changes to the original data mining model resulting from the application of the at least one command to the original data mining model at the server are transmitted from the server to the client;

wherein the document is a mark-up language document; and,

wherein each of the original data mining model and the modified data mining model is a cluster model.

5. A server system coupled to a client system for modifying an original data mining model for display at the client system, the original data mining model maintained by the server system, the server system comprising:

a processor coupled to memory and an interface to a network, and adapted to:

receive a request message at the server system from the client system to modify the original data mining model for use at the client system, wherein the request message is generated at the client system and includes at least one command for modifying the original data mining model for use at the client system;

generate change definitions at the server system from the request message to modify the original data mining model for use at the client system, wherein the change definitions define only changes to the original data mining model resulting from application of the at least one command to the original data mining model at the server system;

store the change definitions in a document at the server system; and,

transmit the document from the server system to the client system, wherein the client system generates a modified data mining model to replace the original data mining model at the client system and for display at the client system using the change definitions received from the server system, whereby application of the at least one command to the original data mining model is performed at the server system rather than at the client system and only changes to the original data mining model resulting from the application of the at least one command to the original data mining model at the server system are transmitted from the server system to the client system;

wherein the document is a mark-up language document; and,

wherein each of the original data mining model and the modified data mining model is a decision tree.

6. The server system of claim 5 wherein the request message is a syntactically extended structured query language command.

7. The server system of claim 5 wherein the request message includes at least one of:

an erase split command to remove a portion of the decision tree;

a find split command to grow a portion of the decision tree;

a goto split command to view an alternative decision tree; and,

a force split command to shape the decision tree in accordance with user specifications.

8. A server system coupled to a client system for modifying an original data mining model for display at the client system, the original data mining model maintained by the server system, the server system comprising:

a processor coupled to memory and an interface to a network, and adapted to:

receive a request message at the server system from the client system to modify the original data mining model for use at the client system, wherein the request message is generated at the client system and includes at least one command for modifying the original data mining model for use at the client system;

generate change definitions at the server system from the request message to modify the original data mining model for use at the client system, wherein the change definitions define only changes to the original data mining model resulting from application of the at least one command to the original data mining model at the server system;

store the change definitions in a document at the server system; and,

transmit the document from the server system to the client system, wherein the client system generates a modified data mining model to replace the original data mining model at the client system and for display at the client system using the change definitions received from the server system, whereby application of the at least one command to the original data mining model is performed at the server system rather than at the client system and only changes to the original data mining model resulting from the application of the at least one command to the original data mining model at the server system are transmitted from the server system to the client system;

wherein the document is a mark-up language document; and,

wherein each of the original data mining model and the modified data mining model is a cluster model.

9. A computer program product having computer executable code for directing a server system coupled to a client system to modify a data mining model for display at the client system, the data mining model maintained by the server system, the computer program product comprising:

code for receiving a request message at the server system from the client system to modify the original data mining model for use at the client system, wherein the request message is generated at the client system and includes at least one command for modifying the original data mining model for use at the client system;

code for generating change definitions at the server system from the request message to modify the original data mining model for use at the client system, wherein the change definitions define only changes to the original data mining model resulting from application of the at least one command to the original data mining model at the server system;

code for storing the change definitions in a document at the server system; and,

code for transmitting the document from the server system to the client system, wherein the client system generates a modified data mining model to replace the original data mining model at the client system and for display at the client system using the change definitions received from the server system, whereby application of the at least one command to the original data mining model is performed at the server system rather than at the client system and only changes to the original data mining model resulting from the application of the at least one command to the original data mining model at the server system are transmitted from the server system to the client system;

wherein the document is a mark-up language document; and,

wherein each of the original data mining model and the modified data mining model is a decision tree.

10. The computer program product of claim 9 wherein the request message is a syntactically extended structured query language message command.

11. The computer program product of claim 9 wherein the request message includes at least one of:

an erase split command to remove a portion of the decision tree;

a find split command to grow a portion of the decision tree;

a goto split command to view an alternative decision tree; and,

a force split command to shape the decision tree in accordance with user specifications.

12. A computer program product having computer executable code for directing a server system coupled to a client system to modify a data mining model for display at the client system, the data mining model maintained by the server system, the computer program product comprising:

code for receiving a request message at the server system from the client system to modify the original data mining model for use at the client system, wherein the request message is generated at the client system and includes at least one command for modifying the original data mining model for use at the client system;

code for generating change definitions at the server system from the request message to modify the original data mining model for use at the client system, wherein the change definitions define only changes to the original data mining model resulting from application of the at least one command to the original data mining model at the server system;

code for storing the change definitions in a document at the server system; and, code for transmitting the document from the server system to the client system, wherein the client system generates a modified data mining model to replace the original data mining model at the client system and for display at the client system using the change definitions received from the server system, whereby application of the at least one command to the original data mining model is performed at the server system rather than at the client system and only changes to the original data mining model resulting from the application of the at least one command to the original data mining model at the server system are transmitted from the server system to the client system;

wherein the document is a mark-up language document; and,

wherein each of the original data mining model and the modified data mining model is a cluster model.