Graphical model for data validation

1. A method of determining whether an element from a source has been properly assigned to a category by the source comprising:

Receiving data from the source;

Based on a training routine, determining a confidence level that a specific element in the data is correctly assigned to a master category given that the source assigned the data to one of the plurality of assigned source categories;

If confidence level meets a threshold, store the element in the assigned master category; and

If the confidence level does not meet a threshold, storing the element for reassignment.

2. The method of claim 1 , wherein the master categories are placed as nodes in a graph and the assigned source categories are related to the nodes in the graph.

3. The method of claim 2 , wherein the graph has nodes and nodes may be parents or children.

4. The method of claim 2 , wherein the nodes of graph are traversed.

5. The method of claim 2 , further comprising if an element has been assigned to a child node, also assigning the element to a parent node.

6. The method of claim 2 , further comprising iterating through the categories in an attempt to assign an element to the most specific category.

7. The method of claim 2 , wherein the confidence level is calculated using a sum product algorithm.

8. The method of claim 2 , further comprising calculating a joint probability density function of all master categories and all source categories.

9. A computer storage medium comprising computer executable code for executing a method of determining whether an element from a source has been properly assigned to a category by the source, the computer code comprising code for:

Receiving data from the source;

Based on a training routine, determining a confidence level that a specific element in the data is correctly assigned to a master category given that the source assigned the data to one of the plurality of assigned source categories;

If confidence level meets a threshold, store the element in the assigned master category; and

If the confidence level does not meet a threshold, storing the element for reassignment.

10. The computer storage medium of claim 9 ,

wherein the master categories are placed as nodes in a graph and the assigned source categories are related to the nodes in the graph;

wherein the graph has nodes and nodes may be parents or children; and

wherein the nodes of graph are traversed.

11. The computer storage medium of claim 9 , further comprising computer executable code for determining the conditional probability that the assignment is correct given the training routine wherein the training routine comprises:

Reviewing a mapping from a training sample to determine if each assigned source category is correct; and

Storing whether the mapping was correct for each assigned source category.

12. The computer storage medium of claim 9 , further comprising computer executable code for assigning the element to a parent node if an element has been assigned to a child node.

13. The computer storage medium of claim 9 , further comprising computer executable code for iterating through the categories in an attempt to assign an element to the most specific category.

14. The computer storage medium of claim 9 , further comprising computer executable code for

using a sum product algorithm to calculate a confidence level, and

calculating a joint probability density function of all master categories and all source categories.

15. A computer system comprising:

a processor configured according to computer executable instructions,

a memory in communication with the processor and

an input output circuit,

the computer executable instructions comprising instruction for executing a method of determining whether an element from a source has been properly assigned to a category by the source, the computer instructions comprising instructions for:

Receiving data from the source;

Based on a training routine, determining a confidence level that a specific element in the data is correctly assigned to a master category given that the source assigned the data to one of the plurality of assigned source categories;

If confidence level meets a threshold, store the element in the assigned master category;

If the confidence level does not meet a threshold, storing the element for reassignment;

wherein the master categories are placed as nodes in a graph and the assigned source categories are related to the nodes in the graph;

wherein the graph has nodes and nodes may be parents or children; and

wherein the nodes of graph are traversed.

16. The computer system of claim 15 , further comprising computer executable instruction for determining the conditional probability that the assignment is correct given the training routine wherein the training routine comprises:

Reviewing a mapping from a training sample to determine if each assigned source category is correct; and

Storing whether the mapping was correct for each assigned source category.

17. The computer system of claim 15 , further comprising computer executable code for assigning the element to a parent node if an element has been assigned to a child node.

18. The computer system of claim 15 , further comprising computer executable code for

using a sum product algorithm to calculate a confidence level, and

calculating a joint probability density function of all master categories and all source categories.