System and method of semantic based searching

Abstract

A computer-implemented method is provided for searching documents containing complex bodies of knowledge, such as patents and research papers. The computer-implemented method and related hardware and software provides methodology to interpret the intent of the searcher (the meaning of the searcher's query) into a MetaLanguage, including but not limited to the use of Fundamental Nature Attributes, Fundamental Action Attributes and Weighting of these attributes as it pertains to the intent of the searcher. The invention relates to semantic based searches. The same methodology that is used on the searcher's query is also used to mine and store the existing databases of patents and research papers into databases of MetaLanguage for the purpose of producing search results that better match search inquiries.

What is claimed is: 
     
         1 . A computer implemented method of searching, comprising the steps of:
 selecting a target set of searchable data items;   processing each of said data items through a MetaLanguage translator, said translator performing the steps of:
 identifying relationships and qualifiers of patterns contained within each data item; 
 assigning at least one characteristic to each of said relationships, wherein said characteristic is a single abstraction of more than one pattern or qualifier of a pattern; and 
 associating in a computer searchable memory said at least one characteristic with the data item to which it relates; 
   inputting a search query;   assigning at least one characteristic to said search query; and   searching said computer searchable memory to identify any data items that include at least one characteristic associated therewith that is relevant to said at least one characteristic of said search query.   
     
     
         2 . The method as claimed in  claim 1  further comprising the steps of:
 generating a results list of data items identified in said searching step; and 
 displaying said results list on a graphical user interface. 
 
     
     
         3 . The method as claimed in  claim 1  wherein said at least one characteristic is a Fundamental Attribute that includes a Fundamental Nature and a Fundamental Action. 
     
     
         4 . The method as claimed in  claim 3  wherein said at least one Fundamental Attribute further includes a Fundamental Qualifier. 
     
     
         5 . The method as claimed in  claim 1  wherein separate sections of each item of said target set of searchable data items are identified in said indentifying step and wherein at least one characteristic is assigned to each section of each item in said assigning step. 
     
     
         6 . The method as claimed in  claim 5  where said separate sections comprise individual sentences within each item. 
     
     
         7 . The method as claimed in  claim 5  wherein said separate sections comprise individual pages within each item. 
     
     
         8 . The method as claimed in  claim 5  wherein said separate sections comprise the entirety of each item. 
     
     
         9 . The method as claimed in  claim 5  wherein multiple characteristics for a single section of each item are associated together to determine at least one characteristic for said section. 
     
     
         10 . The method as claimed in  claim 1  wherein in said searching step said MetaLanguage translator determines that said at least one characteristic associated with a data item is relevant to said at least one characteristic of said search query through a ranking process. 
     
     
         11 . The method as claimed in  claim 10  wherein said ranking process includes a prioritization of relevance based on similarity of said at least one characteristic associated with a data item to said at least one characteristic of said search query. 
     
     
         12 . The method as claimed in  claim 11  wherein said prioritization is based upon a frequency of associations between said at least one characteristic associated with a data item and said at least one characteristic of said search query. 
     
     
         13 . The method as claimed in  claim 10  wherein said ranking process includes a prioritization of relevance based on pattern recognition. 
     
     
         14 . The method as claimed in  claim 13  wherein said prioritization is based upon user feedback regarding relevance of items identified in said searching step. 
     
     
         15 . The method as claimed in  claim 1  wherein said characteristic are selected from a computer searchable memory in which each characteristic is associated with more than one pattern or qualifier of a pattern. 
     
     
         16 . The method as claimed in  claim 1  wherein in said inputting step, a user is guided into inputting said search query in the form of an act and an outcome desired. 
     
     
         17 . The method as claimed in  claim 1  wherein in said searching step includes a first order search and a second order search option. 
     
     
         18 . The method as claimed in  claim 1  further comprising the step of utilizing crowd sourcing to assess the validity and relevance of results of said searching step. 
     
     
         19 . The method as claimed in  claim 1  wherein in said processing step, said MetaLanguage translator further performs the steps of:
 associating in said computer searchable memory each of said relationships and qualifiers of patterns contained within each data item with the data item to which each of said relationships and qualifiers of patterns contained within each data item relates; 
 identifying relationships and qualifiers of patterns contained within said search query; and 
 searching said computer searchable database memory to identify any data items that include at least one relationship or qualifiers of patterns associated therewith that are relevant to at least one of said relationships or qualifiers of patterns of said search query. 
 
     
     
         20 . The method as claimed in  claim 19  wherein said step of searching said computer searchable database memory to identify any data items that include at least one relationship or qualifiers of patterns associated therewith that are relevant to at least one of said relationships or qualifiers of patterns of said search query is performed prior to said step of searching said computer searchable memory to identify any data items that include at least one characteristic associated therewith that is relevant to said at least one characteristic of said search query. 
     
     
         21 . The method as claimed in  claim 20  wherein said step of searching said computer searchable memory to identify any data items that include at least one characteristic associated therewith that is relevant to said at least one characteristic of said search query is conducted within a results list of said step of searching said computer searchable database memory to identify any data items that include at least one relationship or qualifiers of patterns associated therewith that are relevant to at least one of said relationships or qualifiers of patterns of said search query. 
     
     
         22 . The method as claimed in  claim 19  wherein said step of searching said computer searchable memory to identify any data items that include at least one characteristic associated therewith that is relevant to said at least one characteristic of said search query is performed prior to said step of searching said computer searchable database memory to identify any data items that include at least one relationship or qualifiers of patterns associated therewith that are relevant to at least one of said relationships or qualifiers of patterns of said search query. 
     
     
         23 . The method as claimed in  claim 22  wherein said step of searching said computer searchable memory to identify any data items that include at least one characteristic associated therewith that is relevant to said at least one characteristic of said search query is conducted within a results list of said step of searching said computer searchable database memory to identify any data items that include at least one relationship or qualifiers of patterns associated therewith that are relevant to at least one of said relationships or qualifiers of patterns of said search query.