Back to results
US 11,107,173 B2Granted
Usage data management system and method
AssigneeSoundExchange, Inc.
InventorsEdgar Castilla, Reid Badgett, Christopher B. Yorks
Life of this patent
Estimating the term…
Abstract
A management system and method for usage data from usage logs of works of authorship is characterized by the comparison of data components from the usage logs to similar data components for works of authorship stored in a database. The usage data is normalized and cleaned prior to comparison, and additional information from the usage data can be extracted for later use in the comparison. False positive matches are eliminated from the comparison so that incompatible potential candidate matches are eliminated. In addition, the best match from a group of multiple matches is determined for a particular work from a usage log.
Claims14
A system for managing usage text-based metadata from a usage log of a plurality of works of authorship from a service provider, comprising
- (a) a usage data management system containing a database of text-based authoritative metadata from a plurality of works of authorship, said text-based authoritative metadata including at least one field of text-based metadata corresponding with at least one of an author name, a title, a work identification code, a recording album, a publisher name, and a marketing label of the works of authorship in the database;
- (b) a license management system which contains the usage logs from each of the service providers, said usage logs including text-based usage metadata including at least one field of text-based metadata corresponding with at least one of an author name, a title, a work identification code, a recording album, a publisher name, and a marketing label of each of the works of authorship being reproduced by each service provider;
- and
- (c) an automated matching subsystem arranged within said usage data management system and in communication with said license management system, wherein said automated matching subsystem
- (1) cleans and normalizes the text-based usage metadata;
- (2) compares each field of said text-based usage metadata from the usage logs against each corresponding field of said text-based authoritative metadata from said usage data management system using a separate matching algorithm for each field of text-based metadata to produce a matching identification value for each comparison;
- (3)repeats the comparison step for each field of text-based metadata until a strong candidate match is found;
- (4) sums the highest matching identification values for each field of text-based metadata to obtain a total matching score;
- (5) compares the total matching score with a threshold score for each field of text-based metadata;
- and
- (6) determines a match for each work of authorship on the usage log with a stored work of authorship in the database when at least one total matching score exceeds a threshold score for each field of text-based metadate, thereby to identify each work of authorship from the usage logs.
8 more elements
A method for managing text-based usage metadata from a usage log of a plurality of works of authorship from a service provider, comprising the steps of
- (a) storing, by a usage data management system, text-based authoritative metadata from a plurality of works of authorship in a database in the usage data management system, said text-based authoritative metadata including at least one field of text-based metadata corresponding with at least one of an author name, a title, a work identification code, a recording album, a publisher name, and a marketing label of the works of authorship in the database;
- (b) normalizing, by an automated matching subsystem, the text-based usage metadata from the usage log said text-based usage metadata including at least one field of text-based metadata corresponding with at least one of an author name, a title, a work identification code, a recording album, a publisher name, and a marketing label of each of the works of authorship being reproduced by each service provider;
- (c) extracting, by the automated matching subsystem, a plurality of properties from the normalized text-based usage metadata after said normalizing step, said extracting step identifying additional information from the normalized text-based usage metadata independent of at least one of the author name, the title, the work identification code, the recording album, the publisher name, and the marketing label;
- (d) comparing, by the automated matching subsystem, each field of the normalized text-based usage metadata from the usage log against each corresponding field of previously stored text-based authoritative metadata from the database using a separate matching algorithm for each field of text-based metadata to produce a matching identification value for each comparison;
- (e) repeating said comparison step for each field of text-based metadata until a strong candidate match is found;
- (f) summing the highest matching identification values for each field of text-based metadata to obtain a total matching score;
- (g) comparing said total matching score with a threshold score for each field of text-based metadata;
- and
- (h) determining a match for each work of authorship on the usage log with a stored work of authorship in the database when at least one total matching score exceeds a threshold score for each field of text-based metadata, thereby to identify each of the works of authorship on the usage log.
6 more elements
Dependencies are inferred from the claim text.