Method to compress linguistic structures
AssigneesSony Corporation, Sony Electronics Inc.
InventorEugene Koontz
Life of this patent
Estimating the term…
Abstract
A method and system for compressing a data structure. A segment is identified within the data structure. Each segment identified is counted for the number of occurrences of the segment within the data structure. If the number of occurrences is greater than one, the segment is saved in a recurring data structure. Also, the recurring segment within the data structure is replaced with an index to the segment stored in the recurring data structure.
Claims28
A computerized method for compressing a data structure, comprising:
- identifying at least one segment within the data structure, wherein the at least one segment is a discrete sub-structure of the data structure and comprises at least one feature-value pair;
- and
- if the number of occurrences of the at least one segment within the data structure is greater than one,
- saving the at least one segment in a recurring data structure, and
- replacing the at least one segment within the data structure with an index corresponding to the at least one segment in the recurring data structure.
A computerized system for compressing a data structure, comprising:
- a segmentator configured to determine at least one segment within the data structure, wherein the at least one segment is a discrete sub-structure of the data structure and comprises at least one feature-value pair;
- a segment counter configured to count a number of occurrences of the at least one segment within the data structure;
- and
- a recurring data structure for storing the at least one segment if the number of occurrences of the at least one segment is greater than one, wherein the at least one segment in the data structure is replaced by an index corresponding to the at least one segment in the recurring data structure.
An article of manufacture including one or more computer-readable media with executable instructions therein, which, when executed by a processing device causes the processing device to:
- identify at least one segment within the data structure, wherein the at least one segment is a discrete sub-structure of the data structure and comprises at least one feature-value pair;
- count a number of occurrences of the at least one segment within the data structure;
- save the at least one segment in a recurring data structure if the number of occurrences is greater than one;
- and
- replace the at least one segment within the data structure with an index corresponding to the at least one segment in the recurring data structure.
An apparatus for compressing a data structure comprising:
- means for identifying at least one segment within the data structure, wherein the at least one segment is a discrete sub-structure of the data structure and comprises at least one feature-value pair;
- means for counting a number of occurrences of the at least one segment within the data structure;
- and
- a recurring data structure for storing the at least one segment if the number of occurrences of the at least one segment is greater than one, wherein the at least one segment in the data structure is replaced by an index corresponding to the at least one segment in the recurring data structure.
Dependencies are inferred from the claim text.