Data model optimization

Abstract

A name of one or more entity classes of the data model may be refined to conform to a naming convention. A semantic meaning of each of the names and one or more attributes of each entity class may be determined. It may be determined that the name of a first entity class is semantically similar to the name of a second entity class based on a semantic distance between the semantic meaning of the names, where a substantial similarity may be determined between the first entity class and the second entity class by comparing the semantic meaning of the one or more attributes of the first entity class to the semantic meaning of the one or more attributes of the second entity class. The data model may be normalized based on the substantial similarity.

What is claimed is: 
     
         1 . A computer-implemented method for refining a data model having a plurality of
 entity classes, the computer-implemented method comprising:   comparing a semantic meaning of a name of a first entity class to a semantic meaning of a second entity class;   comparing a semantic meaning of an attribute of the first entity class to a semantic meaning of an attribute of the second entity class;   determining that a substantial similarity exists between the first entity class and the second entity class based on the comparison of the semantic meaning of the names of the first and second entity classes and the comparison of the semantic meaning of the attributes of the first and second entity classes; and   collapsing the first entity class into the second entity class based on the substantial similarity between the first entity class and the second entity class.   
     
     
         2 . The computer-implemented method of  claim 1  further comprising refining the names of each of the first and second entity classes to conform to a core component technical specification (CCTS) naming convention. 
     
     
         3 . The computer-implemented method of  claim 1  further comprising identifying, for the names of the first and second entity classes, a dictionary entry name (DEN) including an object term, a property term, and a representation term. 
     
     
         4 . The computer-implemented method of  claim 1  wherein the comparing the semantic meanings of the names and attributes comprises determining the semantic meanings of the names and attributes based on a single dictionary including synonym sets. 
     
     
         5 . The computer-implemented method of  claim 1  wherein the comparing the semantic meaning of the name of the first entity class to the name of the second entity class comprises:
 determining a first semantic distance between the semantic meaning of the name of the first entity class and a root entity class from the first and second entity classes; 
 determining a second semantic distance between the semantic meaning of the name of the second entity class and the root entity class; and 
 comparing the first semantic distance to the second semantic distance to determine a third semantic distance between the semantic meaning of the name of the first entity class and the semantic meaning of the name of the second entity class. 
 
     
     
         6 . The computer-implemented method of  claim 1  wherein the determining that substantial similarity exists between the name of the first entity class and the name of the second entity class comprises determining that semantic distances between the names and attributes are less than a semantic threshold distance. 
     
     
         7 . The computer-implemented method of  claim 1  wherein the determining that the substantial similarity between the first entity class and the second entity class comprises:
 determining a first semantic meaning of the attribute of the first entity class; 
 determining a second semantic meaning of the attribute of the second entity class; and 
 comparing the first semantic meaning with the second semantic meaning to determine whether the substantial similarity exists. 
 
     
     
         8 . The computer-implemented method of  claim 7  wherein the comparing the first semantic meaning with the second semantic meaning comprises:
 determining an attribute semantic distance between the first semantic meaning and the second semantic meaning; and 
 determining that the attribute semantic distance is less than a semantic threshold distance. 
 
     
     
         9 . The computer-implemented method of  claim 1  further comprising creating a new entity class that includes the attribute of the first entity class and the attribute of the second entity class. 
     
     
         10 . The computer-implemented method of  claim 1  wherein the collapsing comprises:
 adding at least a portion of the first entity class to the second entity class based on the substantial similarity of the attribute of the first entity class and the attribute of the second entity class; and 
 removing the first entity class based on the substantial similarity between the first entity class and the second entity class. 
 
     
     
         11 . The computer-implemented method of  claim 1  further comprising:
 providing a message to a user requesting authorization to perform refining; 
 receiving the requested authorization; and 
 refining the names of each of the first entity class and the second entity class to conform to a naming convention based on receiving the requested authorization. 
 
     
     
         12 . The computer-implemented method of  claim 1  further comprising:
 providing a message to a user requesting authorization to perform the collapsing; 
 receiving the requested authorization; and 
 performing the collapsing based on receiving the requested authorization. 
 
     
     
         13 . A system for refining a data model having a plurality of entity classes, the system comprising:
 a semantics analyzer configured to determine semantic meanings of names and attributes of the plurality of entity classes of the data model; and   a comparator configured to:
 compare a semantic meaning of a name of a first entity class to a semantic meaning of a second entity class; 
 compare a semantic meaning of an attribute of the first entity class to a semantic meaning of an attribute of the second entity class; 
 determine that a substantial similarity exists between the first entity class and the second entity class based on the comparison of the semantic meaning of the names of the first and second entity classes and the comparison of the semantic meaning of the attributes of the first and second entity classes; and 
 collapse the first entity class into the second entity class based on the substantial similarity between the first entity class and the second entity class. 
   
     
     
         14 . The system of  claim 13 , wherein the comparator is configured to compare the semantic meaning of the name of the first entity class to the name of the second entity class by
 determining a first semantic distance between the semantic meaning of the name of the first entity class and a root entity class from the first and second entity classes,   determining a second semantic distance between the semantic meaning of the name of the second entity class and the root entity class, and   comparing the first semantic distance to the second semantic distance to determine a third semantic distance between the semantic meaning of the name of the first entity class and the semantic meaning of the name of the second entity class.   
     
     
         15 . The system of  claim 13 , wherein the comparator is configured to collapse the first entity class into the second entity class by
 adding at least a portion of the first entity class to the second entity class based on the substantial similarity of the attribute of the first entity class and the attribute of the second entity class, and   removing the first entity class based on the substantial similarity between the first entity class and the second entity class.   
     
     
         16 . A computer-implemented method for normalizing a first entity class of a data model having the first entity class and a second entity class, wherein each of the first and second entity classes includes a name and first and second attributes, the computer-implemented method comprising:
 determining that the first and second attributes of the second entity class are not substantially similar to either the first or second attributes of the first entity class;   for each of the first and second attributes of the first entity class:
 determining a first medial distance between the attribute of the first entity class and the name of the first entity class; and 
 determining a second medial distance between the attribute of the first entity class and the name of the second entity class, and 
   modifying the first entity class by moving one of the first and second attributes of the first entity class to the second entity class so that, for the attribute remaining in the first entity class, the first medial distance between the remaining attribute and the name of the first entity class is less than the second medial distance between the remaining attribute and the name of the second entity class.   
     
     
         17 . The computer-implemented method of  claim 16  further comprising:
 determining that the first and second attributes of the first entity class each include a non-unique property qualifier; and 
 creating a new entity class based on the non-unique property qualifier and including the first and second attributes of the first entity class. 
 
     
     
         18 . The computer-implemented method of  claim 16  wherein:
 determining the first medial distance comprises determining a first semantic distance between the attribute of the first entity class and the name of the first entity class; and 
 determining the second medial distance comprises determining a second semantic distance between the attribute of the first entity class and the name of the second entity class. 
 
     
     
         19 . The computer-implemented method of  claim 16  wherein a dependent relationship exists between the first entity class and the second entity class. 
     
     
         20 . The computer-implemented method of  claim 16  further comprising recursively performing the computer-implemented method for the second entity class and a third entity class of the data model, the third entity class including a name and first and second attributes.