Back to results
US 2006/0149565 A1Published application
Local item extraction
AssigneeGoogle Inc.
InventorMichael Riley
Life of this patent
Estimating the term…
Abstract
A system identifies a document that includes an address and locates business information in the document. The system assigns a confidence score to the business information, where the confidence score relates to a probability that the business information is associated with the address. The system determines whether to associate the business information with the address based on the assigned confidence score.
Claims42
A method, comprising:
- identifying a document that includes an address;
- locating business information in the document;
- assigning a confidence score to the business information, the confidence score relating to a probability that the business information is associated with the address;
- and
- determining whether to associate the business information with the address based on the assigned confidence score.
2 more elements
A system, comprising:
- means for identifying a document that includes an address;
- means for locating one or more business information candidates in the document;
- means for assigning a confidence score to each of the one or more business information candidates, the confidence score associated with one of the business information candidates relating to a probability that the business information candidate is associated with the address;
- and
- means for determining whether to associate one of the one or more business information candidates with the address based on the assigned confidence score.
2 more elements
A system, comprising:
- a memory to store a statistical model;
- and
- a processor, connected to the memory, to:
- identify a document that includes an address,
- identify business information in the document,
- predict whether the business information is associated with the address based on the statistical model, and
- determine whether to associate the business information with the address based on the prediction.
4 more elements
A method, comprising:
- identifying a document that includes an address;
- identifying a plurality of terms that precede the address in the document;
- determining a probability that each of the terms is part of a title associated with the address;
- identifying a candidate title based on one or more of the terms that have a high probability of being part of a title associated with the address;
- assigning a confidence score to the candidate title;
- and
- determining whether to associate the candidate title with the address based on the assigned confidence score.
4 more elements
A method, comprising:
- identifying a document that includes an address;
- identifying a set of candidate telephone numbers in the document;
- determining a probability that each candidate telephone number in the set of candidate telephone numbers is associated with the address;
- and
- determining whether to associate one of the candidate telephone numbers with the address based on the determined probability.
2 more elements
A method, comprising:
- identifying a web page that includes a landmark;
- identifying an attribute in the web page;
- assigning a confidence score to the attribute, the confidence score relating to a probability that the attribute is associated with the landmark;
- and
- determining whether to associate the attribute with the landmark based on the assigned confidence score.
2 more elements
Dependencies are inferred from the claim text.