Method to detect and score churn in online social games
View Patent ↗A method and a system for predicting churn of a player of an online game is described. Online engagements of a group of players of the online game are monitored during a churn prediction model training period. Online engagement scores for the group of players are computed within a periodic number of days within the churn model training period. A weighted exponential moving average of the online engagement scores of the group of players of the online game is computed during the churn model training period. The weighted exponential moving average is used to determine an online engagement threshold value of a churn prediction model for the online game. The online engagement threshold value is applied to a weighted exponential moving average of a player during an observation period to determine a churn probability of the player within a prediction period.
1. A computer-implemented method comprising:
monitoring online engagements of a group of players of an online game during a training period;
using at least one processor to compute respective online engagement scores for the
group of players based on how many days each respective player has played the online game within a subset of the training period;
assigning respective exponential weights to the online engagement scores so that the respective exponential weights of the online engagement scores increase with an increase in recency within the training period;
using the exponential weights to compute a group moving average, the group moving average comprising a weighted exponential moving group average of the online engagement scores of the group of players of the online game during the training period;
using the group moving average to determine an online engagement threshold value of a churn prediction model for the online game; and
applying the online engagement threshold value to a player moving average to determine a churn probability of a player for a prediction period outside the training period, the player moving average comprising a weighted exponential moving player average of an online engagement score of the player during an observation period within the training period.
2. The computer-implemented method of claim 1 , wherein the observation period is shorter than the training period.
3. The computer-implemented method of claim 1 , wherein the online engagement scores for the group of players are based on respective numbers of days per week the group of players played on the online game during the training period.
4. The computer-implemented method of claim 1 , further comprising:
generating a message to the player to keep the player engaged in the online game when the churn probability of the player exceeds a predetermined churn probability threshold.
5. The computer-implemented method of claim 1 , further comprising:
generating a message to the player to request the player to complete a task in the online game when the churn probability of the player exceeds a predetermined churn probability threshold.
6. The computer-implemented method of claim 1 , further comprising: identifying another player connected to the player via a social network to help the player complete a task in the online game when the churn probability of the player exceeds a predetermined churn probability threshold; and generating a message to the player to request help from said another player to complete the task in the online game.
7. The computer-implemented method of claim 1 , further comprising:
generating a message to the player to promote another online game when the churn probability of the player exceeds a predetermined churn probability threshold.
8. An apparatus comprising:
an online engagement monitoring module configured to monitor online engagements of a group of players of an online game during a training period;
an online engagement score module configured to compute respective online engagement scores for the group of players based on how many days each respective player has played the online game within a subset of the training period;
a moving average computation module configured to assign respective exponential weights to the online engagement scores so that the respective exponential weights of the online engagement scores increase with an increase in recency within the training period, and to use the exponential weights to compute a group moving average, the group moving average comprising a weighted exponential moving average of the online engagement scores of the group of players of the online game during the training period;
a churn prediction module configured to use the group moving average to determine an online engagement threshold value of a churn prediction model for the online game;
a churn probability module configured to apply the online engagement threshold value to a player moving average to determine a churn probability of a player for a prediction period outside the training period, the player moving average comprising a weighted exponential moving˜average of an online engagement score of the player during an observation period within the training period; and
a storage device coupled to the churn probability module, the storage device configured to store the group moving average, the player moving average, and the churn probability of the player for the prediction period.
9. The apparatus of claim 8 , wherein the observation period is shorter than the training period.
10. The apparatus of claim 8 , wherein the online engagement scores for the group of players are based on respective numbers of days per week the group of players played on the online game during the training period.
11. The apparatus of claim 8 , further comprising:
a recommendation module configured to generate a message to the player to keep the player engaged in the online game when the churn probability of the player exceeds a predetermined churn probability threshold.
12. The apparatus of claim 8 , further comprising:
a recommendation module configured to generate a message to the player to request the player to complete a task in the online game when the churn probability of the player exceeds a predetermined churn probability threshold.
13. The apparatus of claim 8 , further comprising: a recommendation module configured to identify another player connected to the player via a social network to help the player complete a task in the online game when the churn probability of the player exceeds a predetermined churn probability threshold, and to generate a message to the player to request help from said another player to complete the task in the online game.
14. The apparatus of claim 8 , further comprising:
a recommendation module configured to generate a message to the player to promote another online game when the churn probability of the player exceeds a predetermined churn probability threshold.
15. A non-transitory computer-readable storage medium storing a set of instructions that, when executed by a processor, cause the processor to perform operations, comprising:
monitoring online engagements of a group of players of an online game during a training period
computing respective online engagement scores for the group of players based on how many days each respective player has played the online game within a subset of the training period;
assigning respective exponential weights to the online engagement scores so that the respective exponential weights of the online engagement scores increase with an increase in recency within the using training period;
using the exponential weights to compute a group moving average, the group moving average comprising a weighted exponential moving group average of the online engagement scores of the group of players of the online game during the training period;
using the group moving average to determine an online engagement threshold value of a churn prediction model for the online game; and
applying the online engagement threshold value to a player moving average to determine a churn probability of a player for a prediction period outside the training period, the player moving average comprising a weighted exponential moving˜average of an online engagement score of the player during an observation period within the training period.
16. The non-transitory computer-readable storage medium of claim 15 , wherein the observation period is shorter than the training period.
17. The non-transitory computer-readable storage medium of claim 15 , wherein the online engagement scores for the group of players are based on respective numbers of days per week the group of players played on the online game during the training period.
18. The non-transitory computer-readable storage medium of claim 15 , further comprising:
generating a message to the player to promote another online game when the churn probability of the player exceeds a predetermined churn probability threshold.
19. The non-transitory computer-readable storage medium of claim 15 , further comprising:
generating a message to the player to request the player to complete a task in the online game when the churn probability of the player exceeds a predetermined churn probability threshold.
20. The non-transitory computer-readable storage medium of claim 15 , further comprising: identifying another player connected to the player via a social network to help the player complete a task in the online game when the churn probability of the player exceeds a predetermined churn probability threshold; and generating a message to the player to request help from said another player to complete the task in the online game.