Method and software for extracting chemical data

Abstract

Preferred embodiments of the present invention comprise methods and software for processing text documents and extracting chemical data therein. Preferred method embodiments comprise: (a) identifying and tagging one or more chemical compounds within a text document; (b) identifying and tagging physical properties related to one or more of those compounds; (c) translating one or more of those compounds into a chemical structure; (d) identifying and tagging one or more chemical reaction descriptions within the text document; and (e) extracting at least some of the tagged information and storing it in a database.

1 . A method for processing text documents and extracting chemical data therein, comprising:
 identifying and tagging one or more chemical compounds within a text document;   identifying and tagging physical properties related to one or more of said compounds;   translating one or more of said compounds into a chemical structure;   identifying and tagging one or more chemical reaction descriptions within said text document; and   extracting at least some of said tagged information and storing said extracted tagged information in a database.   
     
     
         2 . The method of  claim 1 , wherein at least some of said chemical compounds are described by their names. 
     
     
         3 . The method of  claim 1 , wherein at least some of said chemical compounds are described by molecular formulas. 
     
     
         4 . The method of  claim 1 , wherein said step of identifying and tagging one or more chemical compounds within a text document comprises comparison to a dictionary of chemical name fragments. 
     
     
         5 . The method of  claim 1 , wherein said chemical structure is represented by a connection table. 
     
     
         6 . The method of  claim 1 , further comprising identifying and tagging a specific rule selected from a set of pre-defined rules for an identified and tagged compound within an identified and tagged reaction. 
     
     
         7 . The method of  claim 6 , wherein rules are defined in terms of concepts. 
     
     
         8 . The method of  claim 6 , wherein rules comprise: starting material, reagent, solvent, catalyst, and product. 
     
     
         9 . The method of  claim 1 , further comprising identifying a yield of product within a reaction. 
     
     
         10 . The method of  claim 1 , further comprising translating said extracted tagged information into a format convenient for storing said extracted tagged information in said database. 
     
     
         11 . The method of  claim 1 , further comprising identifying and tagging atomistic properties within said text document. 
     
     
         12 . The method of  claim 11 , wherein said atomistic properties comprise one or more of: molecular formulas, numbers, ranges of numbers, physical values, labels, and references within text. 
     
     
         13 . The method of  claim 1 , wherein said text document is an XML-type document. 
     
     
         14 . The method of  claim 1 , wherein said text document is not an XML-type document. 
     
     
         15 . The method of  claim 14 , further comprising analyzing text based on line breaks, numbering schemes, and special keywords.