Recommendation system forward simulator
Disclosed herein are system, apparatus, article of manufacture, method and/or computer program product embodiments, and/or combinations and sub-combinations thereof, for utilizing a content acquisition recommendation system to generating a set of candidate content assets, generate embeddings and popularity score estimates for the set of candidate content assets, aggregate the set of candidate content assets with a set of existing content assets to generate a simulation set of content assets, determine a target set of users for the simulation set of content assets, generate, for at least a portion of the target set of users and based on a trained machine learning model, a result set of recommended content assets, determining an impact of the candidate content assets located in the result set of recommended content assets and generate a proposal for an acquisition of candidate content assets.
1 . A computer implemented method for content asset acquisition, the method comprising:
generating a set of candidate content assets;
generating embeddings for the set of candidate content assets;
generating popularity score estimates for the set of candidate content assets;
aggregating the set of candidate content assets with a set of existing content assets to generate a simulation set of content assets;
determining a target set of users for the simulation set of content assets;
generating, based on a forward simulation of predicted streaming within the simulation set of content assets and for at least a portion of the target set of users, a result set of recommended content assets;
determining an impact of the candidate content assets located in the result set of recommended content assets, wherein the impact is based on a comparison of at least one streaming metric of the result set to a corresponding metric of the set of existing content assets, and wherein the determining the impact further comprises predicting an increase or decrease of predicted streaming times based on the aggregating of the set of candidate assets with the set of existing content assets; and
generating, based on the impact, a proposal for the content asset acquisition of the candidate content assets.
2 . The method of claim 1 , wherein the generating embeddings for the set of candidate content assets further comprises generating embeddings, for individual content assets within the set of candidate content assets, by one or more of:
confirming that one or more metadata is associated with the individual content assets within the set of candidate content assets;
searching content sources for the one or more metadata associated with the individual content assets within the set of candidate content assets; or
implementing a trained embedding machine learning model to estimate the one or more metadata associated with the individual content assets within the set of candidate content assets.
3 . The method of claim 1 , wherein the generating popularity score estimates for the set of candidate content assets further comprises generating the popularity score estimates, for individual content assets within the set of candidate content assets, by any of:
confirming that one or more popularity score data is associated with the individual content assets within the set of candidate content assets;
searching content sources for the one or more popularity score data associated with the individual content assets within the set of candidate content assets; or
implementing a trained popularity score estimate machine learning model to estimate the one or more popularity score data associated with the individual content assets within the set of candidate content assets.
4 . The method of claim 1 , wherein the aggregating the set of candidate content assets with the set of existing content assets to generate the simulation set of content assets further comprises: associating the embeddings and the popularity score estimates with a corresponding individual candidate content asset within the set of candidate content assets.
5 . The method of claim 1 , further comprising selecting a trained machine learning model to estimate metadata associated with the target set of users, wherein the metadata includes any of: demographics, user interests, user streaming habits, content popularity, content format, content arrangement, or streaming metrics.
6 . The method of claim 1 , wherein the at least a portion of the target set of users is based on a sampling of X percent of the users.
7 . The method of claim 1 , further comprising extracting a rank of each of the candidate content assets located in the result set of recommended content assets and ranking an acquisition position for each of the candidate content assets.
8 . The method of claim 7 , further comprising removing the candidate content assets located in the result set of recommended content assets from the proposal based on their ranking.
9 . The method of claim 7 , further comprising removing the candidate content assets located in the result set of recommended content assets from the set of candidate content assets based on their ranking and generating a new result set of recommended content assets.
10 . The method of claim 1 , wherein the impact comprises any of:
likelihood of a candidate content asset selection;
user reach;
new user traction; or
number of active content streamers.
11 . A system, comprising:
a memory; and
at least one processor coupled to the memory and configured to perform operations comprising:
generating a set of candidate content assets for acquisition by a content distribution ecosystem;
generating embeddings for the set of candidate content assets;
generating popularity score estimates for the set of candidate content assets;
aggregating the set of candidate content assets with a set of existing content assets to generate a simulation set of content assets;
determining a target set of users for the simulation set of content assets;
generating, based on a forward simulation of predicted streaming within the simulation set and for at least a portion of the target set of users, a result set of recommended content assets;
determining an impact of the candidate content assets located in the result set of recommended content assets, wherein the impact is based on a comparison of at least one streaming metric of the result set to a corresponding metric of the set of existing content assets, and wherein the determining the impact further comprises predicting an increase or decrease of predicted streaming times based on the aggregating of the set of candidate assets with the set of existing content assets; and
generating, based on the impact, a proposal for the acquisition of the candidate content assets.
12 . The system of claim 11 , wherein the generating embeddings for the set of candidate content assets further comprises generating embeddings, for individual content assets within the set of candidate content assets, by one or more of:
confirming that one or more metadata is associated with the individual content assets within the set of candidate content assets;
searching content sources for the one or more metadata associated with the individual content assets within the set of candidate content assets; or
implementing a trained embedding machine learning model to estimate the one or more metadata associated with the individual content assets within the set of candidate content assets.
13 . The system of claim 11 , wherein the generating popularity score estimates for the set of candidate content assets further comprises generating the popularity score estimates, for individual content assets within the set of candidate content assets, by any of:
confirming that one or more popularity score data is associated with the individual content assets within the set of candidate content assets;
searching content sources for the one or more popularity score data associated with the individual content assets within the set of candidate content assets; or
implementing a trained popularity score estimate machine learning model to estimate the one or more popularity score data associated with the individual content assets within the set of candidate content assets.
14 . The system of claim 11 , wherein the aggregating the set of candidate content assets with the set of existing content assets to generate the simulation set of content assets further comprises: associating the embeddings and the popularity score estimates with a corresponding individual candidate content asset within the set of candidate content assets.
15 . The system of claim 14 , the operations further comprising selecting a trained machine learning model to estimate metadata associated with the target set of users, wherein the metadata includes any of: demographics, user interests, user streaming habits, content popularity, content format, content arrangement, or streaming metrics.
16 . The system of claim 11 , wherein the at least a portion of the target set of users is based on a sampling of X percent of the users.
17 . The system of claim 11 , the operations further comprising:
extracting a rank of each of the candidate content assets located in the result set of recommended content assets and ranking an acquisition position for each of the candidate content assets; and
removing the candidate content assets located in the result set of recommended content assets from the proposal based on their ranking.
18 . A non-transitory computer-readable medium having instructions stored thereon that, when executed by at least one computing device, cause the at least one computing device to perform operations comprising:
generating a set of candidate content assets for acquisition by a content distribution ecosystem;
generating embeddings for the set of candidate content assets;
generating popularity score estimates for the set of candidate content assets;
aggregating the set of candidate content assets with a set of existing content assets to generate a simulation set of content assets;
determining a target set of users for the simulation set of content assets;
generating, based on a forward simulation of predicted streaming within the simulation set and for at least a portion of the target set of users, a result set of recommended content assets;
determining an impact of the candidate content assets located in the result set of recommended content assets, wherein the impact is based on a comparison of at least one streaming metric of the result set to a corresponding metric of the set of existing content assets, and wherein the determining the impact further comprises predicting an increase or decrease of predicted streaming times based on the aggregating of the set of candidate assets with the set of existing content assets; and
generating, based on the impact, a proposal for the acquisition of the candidate content assets.