Sequential decision analysis techniques for e-sports
Aspects of the subject disclosure may include, for example, training a decision scoring model for an e-sport based on historical data for the e-sport. In various embodiments, the decision scoring model may be trained based on the historical data and on metadata associated with the e-sport. Some embodiments can include identifying a plurality of candidate in-game decision sequences based on decision parameters for a gameplay decision of an ongoing gaming session and a gaming session history for the ongoing gaming session. Various embodiments can include applying the decision scoring model to rank the plurality of candidate in-game decision sequences. Other embodiments are disclosed.
1. A device, comprising:
a processing system including a processor; and
a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations, the operations comprising:
determining decision parameters for a gameplay decision of a particular participant of a plurality of participants in an ongoing gaming session of an electronic sport (e-sport), wherein an e-sport coordinator executing on a server device is communicatively coupled with a plurality of e-sport client devices associated with the plurality of participants in the ongoing gaming session of the e-sport;
based on the decision parameters and a gaming session history for the ongoing gaming session, identifying a plurality of candidate in-game decision sequences;
applying a decision scoring model to rank the plurality of candidate in-game decision sequences, resulting in ranked candidate in-game decision sequences, wherein the decision scoring model is trained using historical data obtained over a network from the e-sport coordinator, and wherein the historical data comprises aggregated statistics associated with one or more prior gaming sessions of the e-sport; and
causing the ranked candidate in-game decision sequences to be outputted over the network, wherein outputting of the ranked candidate in-game decision sequences enables the particular participant to assess an extent to which decision-making by the particular participant has aided or hindered efforts by the particular participant to win in the ongoing gaming session of the e-sport.
2. The device of claim 1 , wherein the applying the decision scoring model to rank the plurality of candidate in-game decision sequences includes determining a respective aggregate score for each candidate in-game decision sequence of the plurality of candidate in-game decision sequences.
3. The device of claim 2 , wherein the respective aggregate score for each candidate in-game decision sequence of the plurality of candidate in-game decision sequences is determined based on respective scores for a plurality of decisions within that candidate in-game decision sequence.
4. The device of claim 1 , wherein the aggregated statistics include one or both of:
aggregated player-level statistics associated with the one or more prior gaming sessions of the e-sport; and
aggregated team-level statistics associated with the one or more prior gaming sessions of the e-sport.
5. The device of claim 1 , wherein the operations further comprise training the decision scoring model for the e-sport using the historical data for the e-sport and metadata associated with the e-sport.
6. The device of claim 5 , wherein the operations further comprise using a metadata extraction model to obtain at least a portion of the metadata associated with the e-sport by analyzing one or more video feeds, one or more audio feeds, or a combination of both.
7. The device of claim 1 , wherein the identifying the plurality of candidate in-game decision sequences includes simulating one or more in-game scenarios based on an assumed decision.
8. The device of claim 1 , wherein each candidate in-game decision sequence of the plurality of candidate in-game decision sequences comprises a potential sequence of decisions occurring subsequent to a same point in time.
9. A non-transitory machine-readable medium, comprising executable instructions that, when executed by a processing system including a processor, facilitate performance of operations, the operations comprising:
determining decision parameters for a gameplay decision of a particular participant of a plurality of participants in an ongoing gaming session of an electronic sport (e-sport), wherein an e-sport coordinator executing on a server device is communicatively coupled with a plurality of e-sport client devices associated with the plurality of participants in the ongoing gaming session of the e-sport;
based on the decision parameters and a gaming session history for the ongoing gaming session, identifying a plurality of candidate in-game decision sequences;
applying a decision scoring model to rank the plurality of candidate in-game decision sequences, resulting in ranked candidate in-game decision sequences, wherein the decision scoring model is trained using historical data and metadata obtained over a network from the e-sport coordinator, and wherein the historical data comprises aggregated statistics associated with one or more prior gaming sessions of the e-sport; and
outputting the ranked candidate in-game decision sequences over the network, wherein the outputting the ranked candidate in-game decision sequences enables the particular participant to assess an extent to which decision-making by the particular participant has aided or hindered efforts by the particular participant to win in the ongoing gaming session of the e-sport.
10. The non-transitory machine-readable medium of claim 9 , wherein the operations further comprise extracting at least a portion of the metadata for the e-sport from a video feed.
11. The non-transitory machine-readable medium of claim 9 , wherein the operations further comprise extracting at least a portion of the metadata associated with the e-sport from an audio feed.
12. The non-transitory machine-readable medium of claim 9 , wherein the applying the decision scoring model to rank the plurality of candidate in-game decision sequences includes determining a respective aggregate score for each candidate in-game decision sequence of the plurality of candidate in-game decision sequences.
13. The non-transitory machine-readable medium of claim 12 , wherein the respective aggregate score for each candidate in-game decision sequence of the plurality of candidate in-game decision sequences is determined based on respective scores for a plurality of decisions within that candidate in-game decision sequence.
14. The non-transitory machine-readable medium of claim 9 , wherein the aggregated statistics include aggregated player-level statistics associated with the one or more prior gaming sessions of the e-sport.
15. The non-transitory machine-readable medium of claim 9 , wherein the aggregated statistics include aggregated team-level statistics associated with the one or more prior gaming sessions of the e-sport.
16. A method, comprising:
determining, by a processing system including a processor, decision parameters for a gameplay decision of a particular participant of a plurality of participants in an ongoing gaming session of an electronic sport (e-sport), wherein an e-sport coordinator executing on a server device is communicatively coupled with a plurality of e-sport client devices associated with the plurality of participants in the ongoing gaming session of the e-sport;
based on the decision parameters and a gaming session history for the ongoing gaming session, identifying, by the processing system, a plurality of candidate in-game decision sequences;
applying, by the processing system, a decision scoring model to rank the plurality of candidate in-game decision sequences, resulting in ranked candidate in-game decision sequences, wherein the decision scoring model is trained using historical data obtained over a network from the e-sport coordinator, and wherein the historical data comprises aggregated player-level statistics or aggregated team-level statistics associated with one or more prior gaming sessions of the e-sport; and
outputting the ranked candidate in-game decision sequences over the network, wherein the outputting the ranked candidate in-game decision sequences enables the particular participant to assess an extent to which decision-making by the particular participant has aided or hindered efforts by the particular participant to win in the ongoing gaming session of the e-sport.
17. The method of claim 16 , wherein the identifying the plurality of candidate in-game decision sequences includes simulating one or more in-game scenarios based on an assumed decision.
18. The method of claim 16 , wherein each candidate in-game decision sequence of the plurality of candidate in-game decision sequences comprises a potential sequence of decisions occurring subsequent to a same point in time.
19. The method of claim 16 , wherein the applying the decision scoring model to rank the plurality of candidate in-game decision sequences includes determining a respective aggregate score for each candidate in-game decision sequence of the plurality of candidate in-game decision sequences.
20. The method of claim 19 , wherein the respective aggregate score for each candidate in-game decision sequence of the plurality of candidate in-game decision sequences is determined based on respective scores for a plurality of decisions within that candidate in-game decision sequence.