Method and apparatus for knowledgebase searching

Abstract

A library of query templates and a dictionary that relates keywords to more abstract concepts are first prepared on a computer system. Each template contains one or more typed variables. A query is then generated by entering into the system one or more keywords. Each keyword is abstracted to a concept. Advantageously, each concept may be further refined, for example, by additional abstraction, or by picking one concept from several candidates, or by successive abstraction and rejection of different keywords until an acceptable concept is found. Next, for the concepts that are obtained, the system finds all query templates are then instantiated with those concepts or with the keywords used to form the concepts. The user then selects the most appropriate query from among the instantiated query templates. The invention may be practiced in formulating queries to access any set of information sources. It is particularly useful to use the invention to access distributed, heterogeneous databases which do not have a single standardized vocabulary or structure.

What is claimed is:  
     
         1 . A method of generating a query to one or more information sources comprising the steps of: 
 providing a library of query templates with concepts as variables,    entering one or more keywords,    for each keyword that is entered, abstracting said keyword to a concept,    finding all query templates that match the concepts that are abstracted,    for each query template that matches, instantiating the query template with the concepts that matched the template or with the keywords that were abstracted into concepts that matched the template, and    selecting one or more of the instantiated query templates.    
     
     
         2 . The method of    claim 1    further comprising the step of entering one or more additional keywords after abstracting previously entered keywords, or deleting previously abstracted concepts.  
     
     
         3 . The method of    claim 1    further comprising the step of selecting one concept out of a list of concepts in case of ambiguity in abstracting from keywords to concepts.  
     
     
         4 . The method of    claim 1    further comprising the step of refining or generalizing a concept.  
     
     
         5 . The method of    claim 1    further comprising the step of refining the generated query.  
     
     
         6 . A method for ordering the relevance of queries generated by the method in    claim 1    using one or more of the following ranking criteria: 
 frequency of usage by a user or a user group,  
 modeling of user preferences and user privileges,  
 computational cost of executing the query,  
 cost of accessing the information sources required for executing the query.  
 
     
     
         7 . A method of generating a query to one or more information sources comprising the steps of: 
 providing a library of query templates with semantic types as variables,    entering one or more keywords,    for each keyword that is entered, abstracting said keyword to a concept,    for each concept, categorizing said concept into a semantic type,    finding all query templates that match the semantic types,    for each query template that matches, instantiating the query template with the keywords or concepts that produced the semantic types that matched the template, and    selecting one or more of the instantiated query templates.    
     
     
         8 . The method of    claim 7    further comprising the step of entering one or more additional keywords after abstracting previously entered keywords, or deleting previously abstracted concepts.  
     
     
         9 . The method of    claim 7    further comprising the step of selecting one concept out of a list of concepts in case of ambiguity of the abstraction from keywords to concepts.  
     
     
         10 . The method of    claim 7    further comprising the step of refining or generalizing a concept or semantic type.  
     
     
         11 . The method of    claim 7    further comprising a step of computing supertypes of the semantic types, and finding query templates that match the list of supertypes.  
     
     
         12 . A method for ordering the relevance of queries generated by the method of    claim 11    using one or more of the following ranking criteria: 
 closeness of the supertypes in the query template to the semantic types derived from the entered keywords,  
 relationship of the semantic types in the query template to semantic types in other query templates.  
 
     
     
         13 . The method of    claim 7    further comprising the step of refining the generated query, for example by setting parameters of analytical tools invoked by the query.  
     
     
         14 . A method for ordering the relevance of queries generated by the method in    claim 7    using one or more of the following ranking criteria: 
 frequency of usage by a user or a user group,  
 modeling of user preferences and user privileges,  
 computational cost of executing the query,  
 cost of accessing the information sources required for executing the query (per usage fees, etc.).  
 
     
     
         15 . A method for generating a query to one or more information sources comprising the steps of: 
 providing a library of query templates with semantic types defined in the Unified Medical Language System Semantic Network as variables,    entering one or more keywords,    for each keyword that is entered, abstracting said keyword to a concept defined in the Unified Medical Language System Metathesaurus,    for each concept that is abstracted, categorizing said concept into a semantic type defined in the Unified Medical Language System Semantic Network,    finding all query templates that match said semantic types,    for each query template that matches, instantiating the query templates with the keywords or concepts that produced the semantic types that matched the template, and    selecting one or more of the instantiated query templates.    
     
     
         16 . The method of    claim 15    further comprising the step of entering one or more additional keywords after abstracting previously entered keywords, or deleting previously abstracted concepts.  
     
     
         17 . The method of    claim 15    further comprising the step of selecting one concept out of a list of concepts in case of ambiguity of the abstraction from keywords to concepts.  
     
     
         18 . The method of    claim 15    further comprising the step of refining or generalizing a concept or semantic type.  
     
     
         19 . The method of    claim 15    further comprising a step of computing supertypes of the semantic types, and finding query templates that match the list of supertypes.  
     
     
         20 . A method for ordering the relevance of queries generated by the method of    claim 9    using one or more of the following ranking criteria: 
 closeness of the supertypes in the query template to the semantic types derived from the entered keywords,  
 relationship of the semantic types in the query template to semantic types in other query templates.  
 
     
     
         21 . The method of    claim 15    further comprising the step of refining the generated query, for example by setting parameters of analytical tools invoked by the query.  
     
     
         22 . A method for ordering the relevance of queries generated by the method in    claim 15    using one or more of the following ranking criteria: 
 frequency of usage by a user or a user group,  
 modeling of user preferences and user privileges,  
 computational cost of executing the query,  
 cost of accessing the information sources required for executing the query (per usage fees, etc.).  
 
     
     
         23 . A system for generating a query to one or more information sources comprising: 
 a library of query templates,    a first display screen for entry of one of more keywords,    a second display screen for display of concepts abstracted from an entered keyword,    a third display screen for display of refinements or generalizations of the abstracted concepts,    means for selecting a concept out of a list of concepts, and for deleting concepts,    a fourth display screen for display of one or more queries that relate the concepts, incorporating refinements and generalizations, and    means for selecting one of said queries.