Data classification

Abstract

The invention relates to a computer-implemented method for classifying a set of data values. For each of the data values of the set of data values, a set of one or more terms associated with the respective data value is determined using one or more first knowledge bases. A set of common terms is determined. The set of common terms comprises terms present in more than one of the sets of terms. For each of the common terms, a number of hits for a lookup query against one or more second knowledge data bases is determined. One or more common terms of the set of common terms with the smallest number of hits are determined and a result is returned. The result comprises the one or more common terms with the smallest number of hits as one or more candidate classes for classifying the set of data values.

1 . A computer-implemented method for classifying a set of data values, the method comprising:
 determining, for each of the data values of the set of data values, a set of one or more terms associated with the respective data value using one or more first knowledge bases,   determining a set of common terms, the set of common terms comprising terms present in more than one of the sets of terms,   determining, for each of the common terms, a number of hits for a lookup query against one or more second knowledge data bases,   determining one or more common terms of the set of common terms with the smallest number of hits; and   returning a result comprising the one or more common terms with the smallest number of hits as one or more candidate classes for classifying the set of data values.   
     
     
         2 . The method of  claim 1 , the result exclusively comprising the one or more common terms with the smallest number of hits. 
     
     
         3 . The method of  claim 1 , the determining of the one or more common terms with the smallest number of hits comprising
 determining an order list of common terms, the order list of common terms comprising the common terms of the set of common terms ordered according to their respective numbers of hits.   
     
     
         4 . The method of  claim 3 , the result comprising the ordered list of common terms, the order list representing a candidate hierarchy of candidate classes for classifying the set of data values. 
     
     
         5 . The method of  claim 3 , the result comprising an excerpt from the ordered list of common terms, the excerpt comprising common terms with numbers of hits smaller than a maximum number of hits. 
     
     
         6 . The method of  claim 1 , further extending the number of terms of one or more of the sets of terms, the extending comprising for each term of each set of terms: querying for one or more additional terms, the additional terms being generalizations or synonyms of the respective term,
 if an additional term is found, adding the respective additional term to the respective set of terms.   
     
     
         7 . The method of  claim 6 , one or more of the following being queried for additional terms: an ontology, a taxonomy, a glossary. 
     
     
         8 . The method of  claim 1 , further comprising reducing the number of common terms comprised by the set of common terms, the reducing comprising for each of the common terms: determining whether the common term is a generalization of another one of the common terms comprised by the set of common terms,
 deleting the common term from the set of common terms if it is determined to be a generalization.   
     
     
         9 . The method of  claim 8 , using for the determining whether the common term is a generalization one or more of the following: an ontology, a taxonomy, a glossary. 
     
     
         10 . The method of  claim 8 , the reducing further comprising for each of the common terms in sequence:
 determining whether the common term is a synonym of one of the remaining common terms comprised by the set of common terms,   deleting the common term from the set of common terms if it is determined to be a synonym.   
     
     
         11 . The method of  claim 10 , using for the determining whether the common term is a synonym one or more of the following: an ontology, a taxonomy, a glossary. 
     
     
         12 . The method of  claim 1 , the set of common terms exclusively comprising terms present in at least a minimum number of the sets of terms, the minimum number being larger than one. 
     
     
         13 . The method of  claim 12 , the minimum number being equal to the number of sets of terms. 
     
     
         14 . A computer program product comprising a non-volatile computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code being configured to implement the following for classifying a set of data values, when being executed by a computer system:
 for each of the data values of the set of data values, determining a set of one or more terms associated with the respective data value using one or more first knowledge bases,   determining a set of common terms, the set of common terms comprising terms present in more than one of the sets of terms,   for each of the common terms, determining a number of hits for a lookup query against one or more second knowledge data bases,   determining one or more common terms of the set of common terms with the smallest number of hits,   returning a result comprising the one or more common terms with the smallest number of hits as one or more candidate classes for classifying the set of data values.   
     
     
         15 . The computer program product of  claim 14 , the further being configured to implement reducing the number of common terms comprised by the set of common terms, the reducing comprising for each of the common terms:
 determining whether the common term is a generalization of another one of the common terms comprised by the set of common terms,   deleting the common term from the set of common terms if it is determined to be a generalization.   
     
     
         16 . The computer program product of  claim 15 , the reducing further comprising for each of the common terms in sequence:
 determining whether the common term is a synonym of one of the remaining common terms comprised by the set of common terms,   deleting the common term from the set of common terms if it is determined to be a synonym.   
     
     
         17 . A computer system for classifying a set of data values, computer system being configured to:
 determine, for each of the data values of the set of data values, a set of one or more terms associated with the respective data value using one or more first knowledge bases,   determine a set of common terms, the set of common terms comprising terms present in more than one of the sets of terms,   determine, for each of the common terms, a number of hits for a lookup query against one or more second knowledge data bases,   determine one or more common terms of the set of common terms with the smallest number of hits,   return a result comprising the one or more common terms with the smallest number of hits as one or more candidate classes for classifying the set of data values.   
     
     
         18 . The computer system of  claim 17 , further being configured to reduce the number of common terms comprised by the set of common terms, the reducing comprising for each of the common terms:
 determining whether the common term is a generalization of another one of the common terms comprised by the set of common terms,   deleting the common term from the set of common terms if it is determined to be a generalization.   
     
     
         19 . The computer system of  claim 18 , the reducing further comprising for each of the common terms in sequence:
 determining whether the common term is a synonym of one of the remaining common terms comprised by the set of common terms,   deleting the common term from the set of common terms if it is determined to be a synonym.