Systems and methods for detecting potential communications fraud
AssigneeGoogle Inc.
InventorsAdam Dingle, Niels Provos, Fritz Schneider, Matt Cutts
Life of this patent
Estimating the term…
Abstract
A system includes a data repository and a processing unit. The data repository stores data associated with a corpus of documents hosted on one or more servers. The processing unit analyzes data associated with a suspect document from the corpus of documents. The processing unit further assigns a score, based on the analyzed data, to the suspect document that indicates whether the suspect document is potentially fraudulent.
Claims20
A method performed by one or more server devices, the method comprising:
- determining, by one or more processors of the one or more server devices, that a document in a set of documents requests personal or private information;
- analyzing, by one or more processors of the one or more server devices, based on the determination that the document requests personal or private information, a link structure of the document;
- identifying, by one or more processors of the one or more server devices, one or more documents with link structures similar to the link structure of the document;
- determining, by one or more processors of the one or more server devices, ranks of the one or more documents with link structures similar to the link structure of the document;
- comparing, by one or more processors of the one or more server devices, a rank of the document to the ranks of the one or more documents with link structures similar to the link structure of the document;
- designating, by one or more processors of the one or more server devices, the document as one of trustworthy or untrustworthy based on a result of the comparing;
- and
- storing, in a memory associated with the one or more server devices, information regarding a result of the designating.
A computer-readable memory device containing instructions for controlling at least one processor to perform a method, the instructions comprising:
- one or more instructions that, when executed by the at least one processor, cause the at least one processor to:
- identify a document in a set of documents as being suspect based on the document requesting personal or private information from a user;
- analyze a link structure of the suspect document;
- identify one or more documents with link structures similar to the link structure of the suspect document;
- determine ranks of the one or more documents with link structures similar to the link structure of the suspect document;
- compare a rank of the suspect document to the ranks of the one or more documents with link structures similar to the link structure of the suspect document;
- and
- assign a score, based on a result of the comparing, to the suspect document that indicates whether the suspect document is potentially fraudulent.
A system, comprising:
- a processing unit to:
- identify a document, from a corpus of documents hosted on one or more servers, as being suspect based on the document soliciting personal or private information from a user;
- analyze a link structure of the suspect document;
- identify one or more documents with link structures similar to the link structure of the suspect document;
- determine ranks of the one or more documents with link structures similar to the link structure of the suspect document;
- compare a rank of the suspect document to the ranks of the one or more documents with link structures similar to the link structure of the suspect document;
- and
- assign a score, based on a result of the comparison, to the suspect document that indicates whether the suspect document is potentially fraudulent.
Dependencies are inferred from the claim text.