Parametric player modeling for computer-implemented games
A game management system generates a multidimensional parametric player model representative of player behavior by one or more players in a computer game. The parametric player model populated is used in the identification of groups or clusters of players, and/or in at least partly automated configuration of custom game content for behavior consistent with the parametric player model. A single parametric player model is defines a single set of parametric values corresponding to multiple predefined gameplay parameters, and can be used to model the behavior of a single respective player or to model the behavior of multiple players (e.g., based on cumulative historical gameplay data for the relevant players), providing a representative player model for those players. The player model is ingested by a content generator configured to generate game content customized to the behavior represented by the player model.
1. A method comprising:
for a player set comprising multiple players of a computer-implemented game, compiling for each player a respective multidimensional player model representing gameplay behavior in the game, the player model comprising a multi-value set of parametric values corresponding to a predefined parameter set defined by multiple gameplay parameters, each parametric value being a numerical value representing player behavior for a respectively corresponding gameplay parameter in the parameter set, so that the set of parametric values for each player model represent a datapoint in a multidimensional parameter space defined by the multiple gameplay parameters as respective dimensions;
calculating distances in the multidimensional parameter space between datapoints in the parameter space represented by the respective player models of the player set;
in an automated procedure that is performed using one or more computer processor devices configured therefor and that is based at least on the calculated distances between respective player models in the multidimensional parameter space, identifying a plurality of player clusters that form respective subsets of the player set;
identifying a representative player model for a particular one of the plurality of player clusters, the representative player model defined by a respective set of parametric values for the predefined parameter set; and
in an at least partly automated procedure using the representative player model as an input, generating custom game content of the game for gameplay consistent with the representative player model.
2. The method of claim 1 , further comprising:
accessing historical gameplay data for the players in the player set, the respective player models being compiled based on the historical gameplay data, wherein the representative player model for the particular player cluster is based at least in part on the player models of the subset of players in the particular player cluster.
3. The method of claim 1 , wherein each parametric value is a real number, so that the parameter space is a continuous multidimensional space.
4. The method of claim 1 , wherein the identifying of the plurality of player clusters comprises proximity clustering of the datapoints based on the calculated distances.
5. The method of claim 1 , wherein the identifying of the representative player model comprises compiling the representative player model based on the respective player models of the particular player cluster.
6. The method of claim 5 , wherein the representative player model for the particular player cluster is compiled by calculating a respective parametric value for each of the gameplay parameters based on respective parametric values of the corresponding gameplay parameters in the player models of the cluster.
7. The method of claim 1 , wherein the generating of custom game content comprises, in an at least partly automated procedure, generating configuration data for a unit of gameplay based at least in part on the representative player model, such that different configuration data is generated in different instances of customization based on different respective representative player models.
8. The method of claim 1 , wherein the generating of custom game content comprises, based at least in part on the representative player model, performing automated evaluation of configuration data for a unit of gameplay.
9. The method of claim 8 , wherein the automated evaluation comprises: simulating gameplay of the unit of gameplay according to the representative player model; and comparing a performance metric for the simulated gameplay to a predefined target metric.
10. A system comprising:
one or more computer processor devices; and
memory storing instructions to configure the system, when the instructions are executed by the one or more computer processor devices, to perform operations comprising:
for a player set comprising multiple players of a computer-implemented game, compiling for each player a respective multidimensional player model representing gameplay behavior in the game, the player model comprising a multi-value set of parametric values corresponding to a predefined parameter set defined by multiple gameplay parameters, each parametric value being a numerical value representing player behavior for a respectively corresponding gameplay parameter in the parameter set, so that the set of parametric values for each player model represent a datapoint in a multidimensional parameter space defined by the multiple gameplay parameters as respective dimensions;
calculating distances in the multidimensional parameter space between datapoints in the parameter space represented by the respective player models of the player set;
in an automated procedure based at least on the calculated distances between respective player models in the multidimensional parameter space, identifying a plurality of player clusters that form respective subsets of the player set;
identifying a representative player model for a particular one of the plurality of player clusters, the representative player model defined by a respective set of parametric values for the predefined parameter set; and
in an at least partly automated procedure using the representative player model as an input, generating custom game content of the game for gameplay consistent with the representative player model.
11. The system of claim 10 , wherein the instructions further configure the system to:
accessing historical gameplay data for the players in the player set, the respective player models being compiled based on the historical gameplay data, wherein the representative player model for the particular player cluster is based at least in part on the player models of the subset of players in the particular player cluster.
12. The system of claim 10 , wherein each parametric value is a real number, so that the parameter space is a continuous multidimensional space.
13. The system of claim 10 , wherein the instructions further configure the system such that the identifying of the plurality of player clusters comprises proximity clustering of the datapoints based on the calculated distances.
14. The system of claim 10 , wherein the instructions further configure the system to:
based on the respective player models of the particular player cluster, compile the representative player model for the particular player cluster.
15. The system of claim 14 , wherein the instructions further configure the system to compile the representative player model for the particular player cluster by calculating a respective parametric value for each of the gameplay parameters based on respective parametric values of the corresponding gameplay parameters in the player models of the cluster.
16. The system of claim 10 , wherein the instructions further configure the system to perform generation of custom game content by, in an at least partly automated procedure, generating configuration data for a unit of gameplay based at least in part on the representative player model, such that different configuration data is generated in different instances of customization based on different respective representative player models.
17. The system of claim 10 , wherein the instructions further configure the system to perform, based at least in part on the representative player model, automated evaluation of configuration data for a unit of gameplay.
18. A non-transitory computer-readable storage medium having stored thereon instructions that, when executed by one or more computer processor devices, cause the one or more computer processor devices to perform operations comprising:
for a player set comprising multiple players of a computer-implemented game, compiling for each player a respective multidimensional player model representing gameplay behavior in the game, the player model comprising a multi-value set of parametric values corresponding to a predefined parameter set defined by multiple gameplay parameters, each parametric value being a numerical value representing player behavior for a respectively corresponding gameplay parameter in the parameter set, so that the set of parametric values for each player model represent a datapoint in a multidimensional parameter space defined by the multiple gameplay parameters as respective dimensions;
calculating distances in the multidimensional parameter space between datapoints in the parameter space represented by the respective player models of the player set;
in an automated procedure that is based at least on the calculated distances between respective player models in the multidimensional parameter space, identifying a plurality of player clusters that form respective subsets of the player set;
identifying a representative player model for a particular one of the plurality of player clusters, the representative player model defined by a respective set of parametric values for the predefined parameter set; and
in an at least partly automated procedure using the representative player model as an input, generating custom game content of the game for gameplay consistent with the representative player model.