Back to results
US 8,914,316 B2Granted
Information similarity and related statistical techniques for use in distributed computing environments
AssigneeYahoo! Inc.
InventorsAshwin Tengli, Aravindan Raghuveer, Krishna Prasad Chitrapura
Life of this patent
Estimating the term…
Abstract
Embodiments of methods, systems and/or apparatuses relating to data processing in distributed computing environments are disclosed. In particular, methods, systems, and/or apparatuses for determining information similarly and/or performing related statistical techniques which may be implemented or operated in a distributed computing environment are disclosed.
Claims22
A method, comprising:
- accessing a record set comprising a plurality of records;
- generating a plurality of terms for said plurality of records, individual ones of said plurality of terms comprising a q-gram of a fixed character length;
- determining a plurality of respective parameter frequencies for said plurality of terms;
- at least partially in response to said determining a plurality of respective parameter frequencies, assigning particular termIDs to said plurality of terms based, at least in part, on said respective parameter frequencies, wherein said particular termIDs comprise an identifier to indicate a term/record frequency within said record set;
- comparing one or more positions of said termIDs for first terms of a first vector of terms for a first candidate record and for second terms of a second vector of terms for a second candidate record to determine whether there is an overlap between the first and second vectors.
2 more elements
A method, comprising:
- accessing a record set comprising a plurality of records;
- generating a plurality of terms for said plurality of records, individual ones of said plurality of terms comprising a q-gram of a fixed character length;
- filtering out terms comprising a suffix of said one or more records;
- determining a plurality of respective parameter frequencies for said plurality of terms in response to said filtering;
- at least partially in response to said determining a plurality of respective parameter frequencies, assigning particular termIDs to said plurality of terms based, at least in part, on said respective parameter frequencies, wherein said particular termIDs comprise an identifier to indicate a term/record frequency within said record set;
- and
- comparing one or more positions of said termIDs for first terms of a first vector of terms for a first candidate record and for second terms of a second vector of terms for a second candidate record to determine whether there is an overlap between the first and second vectors.
4 more elements
Dependencies are inferred from the claim text.