Algorithm for identification of trending content
This application relates to techniques for recommending content to a user of a content distribution system. A server device can generate recommendations as part of a user interface for the content distribution system. The server device can be configured to: calculate a trend score for each of a plurality of digital assets managed by a content distribution system, calculate a recommendation score for a subset of digital assets that are not installed on a client device of a target user, calculate a breakout score for a subset of digital assets managed by the content distribution system each having a cumulative number of downloads below a threshold value, rank the digital assets according to the trend scores, the recommendation scores, or the breakout scores, and generate a visual representation of one or more digital assets to recommend to the user based on the ranking.
1. A method for identifying digital assets to recommend to a user, the method comprising, at a server device:
identifying a set of digital assets within a particular category of digital assets, each digital asset in the set of digital assets having a corresponding breakout date established for the digital asset;
identifying one or more trendsetters associated with the particular category of digital assets;
generating a list of digital assets within the particular category of digital assets downloaded by at least one trendsetter in the one or more trendsetters;
filtering the list of digital assets downloaded by the at least one trendsetter for the particular category of digital assets to exclude digital assets having a cumulative number of downloads above a threshold value;
calculating a respective breakout score for each digital asset in the filtered list of digital assets by counting a number of trendsetters that have downloaded the digital asset;
ranking the filtered list of digital assets by their respective breakout scores to identify breakout content to recommend to a user;
generating a visual representation of at least a subset of the filtered list of digital assets based on the ranking and corresponding breakout dates of the digital assets; and
causing at least one client device to display the visual representation.
2. The method of claim 1 , wherein each genre of music in a plurality of genres of music is associated with a different list of trendsetters for the genre of music.
3. The method of claim 2 , wherein the particular category of digital assets comprises a first genre of music, the method further comprising:
generating a second list of digital assets within a second category of digital assets downloaded by at least one trendsetter in one or more trendsetters associated with the second category of digital assets;
filtering the second list of digital assets downloaded by the at least one trendsetter for the second category of digital assets to exclude digital assets having a cumulative number of downloads above the threshold value; and
calculating a breakout score for each digital asset in the filtered second list of digital assets by counting a number of trendsetters that have downloaded that digital asset.
4. The method of claim 1 , wherein each trendsetter in the one or more trendsetters is a user of a music service that has downloaded at least a threshold number of digital assets prior to a corresponding breakout date for each digital asset in the threshold number of digital assets.
5. The method of claim 1 , wherein a breakout date for a digital asset is identified based on a moving average convergence/divergence (MACD) metric.
6. The method of claim 5 , wherein the MACD metric is calculated as an exponentially weighted moving average (EWMA) of a difference between a short time frame EWMA and a long time frame EWMA of historical download data.
7. The method of claim 6 , wherein the breakout date is identified when the MACD metric increases above a threshold value.
8. The method of claim 1 , further comprising adjusting a ranking of digital assets within the particular category of digital assets based on the breakout scores for each digital asset in the filtered list of digital assets.
9. A non-transitory computer readable storage medium configured to store instructions that, when executed by a processor included in a computing device, cause the computing device to identify digital assets to recommend to a user, by carrying out steps that include:
identifying a set of digital assets within a particular category of digital assets, each digital asset in the set of digital assets having a corresponding breakout date established for the digital asset;
identifying one or more trendsetters associated with the particular category of digital assets;
generating a list of digital assets within the particular category of digital assets downloaded by at least one trendsetter in the one or more trendsetters;
filtering the list of digital assets downloaded by the at least one trendsetter for the particular category of digital assets to exclude digital assets having a cumulative number of downloads above a threshold value;
calculating a respective breakout score for each digital asset in the filtered list of digital assets by counting a number of trendsetters that have downloaded the digital asset;
ranking the filtered list of digital assets by their respective breakout scores to identify breakout content to recommend to the user;
generating a visual representation of at least a subset of the filtered list of digital assets based on the ranking and corresponding breakout dates of the digital assets; and
causing at least one client device to display the visual representation.
10. The non-transitory computer readable storage medium of claim 9 , wherein each genre of music in a plurality of genres of music is associated with a different list of trendsetters for the genre of music.
11. The non-transitory computer readable storage medium of claim 10 , wherein the particular category of digital assets comprises a first genre of music, wherein the steps further include:
generating a second list of digital assets within a second category of digital assets downloaded by at least one trendsetter in one or more trendsetters associated with the second category of digital assets;
filtering the second list of digital assets downloaded by the at least one trendsetter for the second category of digital assets to exclude digital assets having a cumulative number of downloads above the threshold value; and calculating a breakout score for each digital asset in the filtered second list of digital assets by counting a number of trendsetters that have downloaded that digital asset.
12. The non-transitory computer readable storage medium of claim 9 , wherein each trendsetter in the one or more trendsetters is a user of a music service that has downloaded at least a threshold number of digital assets prior to a corresponding breakout date for each digital asset in the threshold number of digital assets.
13. A computing device configured to identify digital assets to recommend to a user, the computing device comprising a processor and a memory including instructions that, when executed by the processor, cause the computing device to perform steps that include:
identifying a set of digital assets within a particular category of digital assets, each digital asset in the set of digital assets having a corresponding breakout date established for the digital asset;
identifying one or more trendsetters associated with the particular category of digital assets;
generating a list of digital assets within the particular category of digital assets downloaded by at least one trendsetter in the one or more trendsetters;
filtering the list of digital assets downloaded by the at least one trendsetter for the particular category of digital assets to exclude digital assets having a cumulative number of downloads above a threshold value;
calculating a respective breakout score for each digital asset in the filtered list of digital assets by counting a number of trendsetters that have downloaded the digital asset;
ranking the filtered list of digital assets by their respective breakout scores to identify breakout content to recommend to the user;
generating a visual representation of at least a subset of the filtered list of digital assets based on the ranking and corresponding breakout dates of the digital assets; and
causing at least one client device to display the visual representation.
14. The computing device of claim 13 , wherein each genre of music in a plurality of genres of music is associated with a different list of trendsetters for the genre of music.
15. The computing device of claim 14 , wherein the particular category of digital assets comprises a first genre of music, wherein the steps further include:
generating a second list of digital assets within a second category of digital assets downloaded by at least one trendsetter in one or more trendsetters associated with the second category of digital assets;
filtering the second list of digital assets downloaded by the at least one trendsetter for the second category of digital assets to exclude digital assets having a cumulative number of downloads above the threshold value; and
calculating a breakout score for each digital asset in the filtered second list of digital assets by counting a number of trendsetters that have downloaded that digital asset.
16. The computing device of claim 13 , wherein each trendsetter in the one or more trendsetters is a user of a music service that has downloaded at least a threshold number of digital assets prior to a corresponding breakout date for each digital asset in the threshold number of digital assets.