Systems and methods for intercepting convergent data streams

1 . A system for directing future activity to diverge from a projected future data stream generated based on past activity, comprising:

one or more processors; and

one or more non-transitory, computer-readable media storing instructions that when executed by the one or more processors cause operations comprising:

determining, based on a first data stream associated with a user, to associate the user with a first cluster that is a type of intervention cluster that indicates that the user is performing one or more activities that require intervention;

generating, based on determining to associate the user with a first cluster, a notification that identifies one or more user activities for the user to execute to exit the first cluster;

in response to receiving a second data stream associated with the user, comparing a first projected future data stream to the second data stream to determine a measure of convergence;

in response to determining that the measure of convergence does not exceed a threshold, generating, based on the second data stream, a second projected future data stream that identifies potential future activities for the user;

processing, using a clustering algorithm, the second projected future data stream to determine that the user is no longer associated with the first cluster; and

in response to determining that the user is no longer associated with the first cluster, generating a notification to the user indicating that the user has successfully exited the first cluster.

2 . A computer-implemented method for directing future activity to diverge from a projected future data stream generated based on past activity, comprising:

in response to receiving a first data stream including a plurality of user activities executed by a user, generating, based on the first data stream, a first projected future data stream including potential future activities for the user;

processing, using a clustering algorithm, the first projected future data stream to determine, from a plurality of clusters, a first cluster to associate with the user, wherein the first cluster is a type of intervention cluster that indicates that the user requires intervention;

in response to determining the first cluster, generating a notification to the user including one or more user activities to execute to exit the first cluster;

determining a measure of convergence based on the first projected future data stream; and

determining whether the measure of convergence exceeds a threshold.

3 . The computer-implemented method of claim 2 , further comprising:

generating, based on a second data stream used to determine the measure of convergence and based on determining that the measure of convergence does not exceed a threshold, a second projected future data stream including potential future activities for the user.

4 . The computer-implemented method of claim 2 ,

wherein determining the measure of convergence based on the first projected future data stream comprises:

comparing the first projected future data stream to a second data stream to determine the measure of convergence,

wherein the first data stream corresponds to a first period of time and the second data stream corresponds to a second period of time, and

wherein the second period of them occurs later than the first period of time.

5 . The computer-implemented method of claim 3 , further comprising:

processing, using a clustering algorithm, the second projected future data stream to determine that the user is no longer associated with the first cluster; and

in response to determining that the user is no longer associated with the first cluster, generating a notification to the user indicating that the user has successfully exited the first cluster.

6 . The computer-implemented method of claim 2 , wherein the type of intervention cluster indicates that the user is executing a negative activity,

wherein the negative activity is an activity executed by the user whose consequence to the user can be mitigated with the one or more user activities.

7 . The computer-implemented method of claim 2 , wherein further comprising:

identifying a plurality of users and a second plurality of user activities, wherein the plurality of users is in the first cluster associated with the user;

identifying notifications that were sent to the plurality of users; and

determining that the one or more user activities were identified by the notifications and mitigated one or more negative consequences.

8 . The computer-implemented method of claim 2 , further comprising:

accessing data from a third-party account,

wherein the third-party account is associated with the user,

wherein the third-party account comprises additional user activities, and

wherein the third-party account can be used for the first data stream or a second data stream.

9 . The computer-implemented method of claim 2 , wherein receiving the first data stream further comprises receiving access to a third-party account associated with the user,

wherein the third-party account comprises a second plurality of user activities executed by the user.

10 . The computer-implemented method of claim 2 , wherein determining the measure of convergence comprises:

processing the first projected future data stream and a second data stream using a model to:

receive, as input, the first projected future data stream and the second data stream;

perform a comparison between the first projected future data stream and the second data stream to identify user activities executed by the comparison of the first projected future data stream and the second data stream; and

generate an output corresponding to the user activities executed by the user associated with the first projected future data stream that is equivalent to potential future activities associated with the second data stream; and

generating the measure of convergence using the output from the model.

11 . The computer-implemented method of claim 2 , wherein the first projected future data stream is generated by:

identifying a plurality of users and a second plurality of user activities, wherein the plurality of users is in the first cluster associated with the user; and

generating the first projected future data stream based on the plurality of users and the second plurality of user activities.

12 . The computer-implemented method of claim 2 , further comprising:

preventing, based on determining that the measure of convergence exceeds the threshold, execution of future activities from the user.

13 . The computer-implemented method of claim 2 , wherein the type of intervention cluster indicates that the user requires direction for future activity.

14 . The computer-implemented method of claim 12 , wherein preventing the execution of the future activities comprises revoking or limiting access to an account associated with the user,

wherein revoking or limiting access to the account includes restricting execution of future activities by the user.

15 . The computer-implemented method of claim 2 , wherein the projected future data stream is a synthetic data stream, and wherein the synthetic data stream is generated by:

identifying a plurality of users and a second plurality of user activities, wherein the plurality of users is in the first cluster associated with the user; and

training a model to generate the synthetic data stream based on the second plurality of user activities corresponding to each user of the plurality of users.

16 . One or more non-transitory, computer-readable media storing instructions that when executed by one or more processors cause operations comprising:

in response to receiving a first data stream based on a plurality of user activities executed by a user, generating, based on the first data stream, a first projected future data stream including potential future activities for the user;

determining to associate the user; with an intervention cluster that indicates that the user is performing at least one activity, of the plurality of user activities, that requires intervention;

generating, based on determining to associate the user with a first cluster, a notification that identifies one or more user activities for the user to execute to exit the first cluster;

comparing the first projected future data stream to a second data stream to determine a measure of convergence; and

determining whether the measure of convergence exceeds a threshold.

17 . The one or more non-transitory, computer-readable media of claim 16 , wherein the intervention includes direction to the user for future activity.

18 . The one or more non-transitory, computer-readable media of claim 16 ,

wherein the first projected future data stream is a synthetic data stream, and

wherein the synthetic data stream is generated by:

training a model to generate the synthetic data stream based on a second plurality of user activities corresponding to a plurality of other users associated with the intervention cluster.

19 . The one or more non-transitory, computer-readable media of claim 16 , wherein the at least one activity is a negative activity whose consequence to the user can be mitigated with another activity of the one or more user activities.

20 . The one or more non-transitory, computer-readable media of claim 16 , wherein the first projected future data stream is generated by:

generating the first projected future data stream based on a plurality of other users associated with the intervention cluster.