System and method for extracting categories of data
AssigneeAncestry.com Operations Inc.
InventorsPeter McCombs, Spencer Koehler, Laryn Brown
Life of this patent
Estimating the term…
Abstract
Lines of data that are from a historical document, such as a digitized city directory, have information extracted and stored in searchable data fields. Words and phrases within the lines of data are identified and tagged. Rendering rules are applied to the tagged words and phrases to extract names, addresses, occupations, spouse information and other data, and store that data in the searchable fields.
Claims14
A method for extracting information from a line of computer readable data and creating a searchable data field in which the extracted information is stored, wherein the line of data includes words representing a name of a person and words representing other information related to the person, the method comprising:
- starting at one end of the line, sequentially capturing each word beginning with a capital letter, until a name delimiter is reached, including:
- identifying words and phrases that appear in the line of data, wherein the words and phrases include:
- Name Words that begin with a capitalized letter;
- a PName Word that is any one of the Name Words in a sequence of Name Words appearing in the line of data, the sequence starting at the one end of the line until a PName Word delimiter is reached, wherein the PName Word delimiter is one of (1) a word that is not a Name Word and (2) a comma;
- and
- a PName Phrase that is sequence of PName Words appearing in the line of data, starting at the one end of the line until a PName Phrase delimiter is reached, wherein the PName Phrase delimiter is any word that is not a PName Word;
- and
- capturing the PName Phrase;
- designating the sequentially captured words as the name of the person;
- storing the designated name in a searchable data field;
- and
- associating the other information from the line of computer readable data with the designated name.
A system for extracting information from a line of computer readable data and creating a searchable data field in which the extracted information is stored, wherein the line of data includes words representing a name of a person and words representing other information related to the person, the system comprising:
- a memory device for storing data is searchable data fields;
- and
- a processor configured to:
- starting at one end of the line of data, sequentially capture each word beginning with a capital letter, until a name delimiter is reached, by:
- identifying words and phrases that appear in the line of data, wherein the words and phrases include:
- Name Words that begin with a capitalized letter;
- a PName Word that is any one of the Name Words in a sequence of Name words appearing in the line of data, the sequence starting at the one end of the line until a PName Word delimiter is reached, wherein the PName Word delimiter is one of (1) a word that is not a Name Word and (2) a comma;
- and
- a PName Phrase that is sequence of PName Words appearing in the line of data, starting at the one end of the line until a PName Phrase delimiter is reached, wherein the PName Phrase delimiter is any word that is not a PName Word;
- and
- capturing the PName Phrase;
- designate the sequentially captured words as the name of the person;
- store the designated name in a searchable data field in the memory device;
- and
- associate the other information from the line of computer readable data with the designated name.
Dependencies are inferred from the claim text.