Generation and exploitation of an approximate language model
First Claim
Patent Images
1. A method comprising:
- executing instructions via a processor of a computing system for;
generating a language model according to an approximate hashing technique, the language model comprising a plurality of event sequences in a target language, each member of the plurality of the event sequences associated with at least one count, the language model comprising a set of data structures organized in a hierarchy with lower levels corresponding to event sequences occurring less frequently being stored using fewer bits, the hierarchy having three or more levels;
querying the language model for a member of the plurality of event sequences; and
determining a probability associated with the member of the plurality of event sequences based on results of the query.
2 Assignments
0 Petitions
Accused Products
Abstract
A system, method, and computer program for generating and exploiting an approximate language model are provided. The method comprises generating a language model according to an approximate hashing technique. The language model comprises a plurality of event sequences in a target language, and each member of the plurality of the event sequences is associated with at least one count. The language model is queried for a member of the plurality of event sequences. A probability associated with the member of the plurality of event sequences is determined based on results of the query.
377 Citations
20 Claims
-
1. A method comprising:
executing instructions via a processor of a computing system for; generating a language model according to an approximate hashing technique, the language model comprising a plurality of event sequences in a target language, each member of the plurality of the event sequences associated with at least one count, the language model comprising a set of data structures organized in a hierarchy with lower levels corresponding to event sequences occurring less frequently being stored using fewer bits, the hierarchy having three or more levels; querying the language model for a member of the plurality of event sequences; and determining a probability associated with the member of the plurality of event sequences based on results of the query. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
10. A system comprising:
-
a processor configured to generate a language model based on a training corpus using an approximate hashing technique, the language model comprising a set of data structures organized in a hierarchy with lower levels corresponding to event sequences occurring less frequently being stored using fewer bits, the hierarchy having three or more levels; and a memory configured to store the language model. - View Dependent Claims (11, 12, 13, 14, 15, 16, 17, 18, 19)
-
-
20. A non-transitory computer readable storage medium having embodied thereon a program, the program being executable by a processor for performing a method for determining a translation, the method comprising:
-
generating a language model according to an approximate hashing technique, the language model comprising a plurality of event sequences in a target language, each member of the plurality of the event sequences associated with at least one count, the language model comprising a set of data structures organized in a hierarchy with lower levels corresponding to event sequences occurring less frequently being stored using fewer bits, the hierarchy having three or more levels; querying the language model for a member of the plurality of event sequences; and determining a probability associated with the member of the plurality of event sequences based on results of the query.
-
Specification