System and method for providing pairings for live digital interactions
A system includes a processor and an interface coupled to the processor. The processor determines that a first user takes a first action, and determines to pair the first user and a second user. The processor pairs the first user and the second user for an interaction, where the interaction lasts for a first time period. The interface receives an indication of a preference of the first user for the second user at a second time. The second time is after the first time period.
1 . A system comprising an interaction-based recommendation tool comprising:
a processor configured to:
determine that a first user takes a first action with respect to an interaction-based recommendation tool;
extract a first set of features from a preference history of the first user, the first set of features indicating preferences of the first user with respect to other users;
extract a second set of features from a preference history of a second user, the second set of features indicating preferences of the second user with respect to other users;
determine to pair the first user and the second user based at least on the first action, the first set of features, and the second set of features;
pair the first user and the second user for an interaction, the interaction lasting for a first time period, the interaction is concurrent for the first user and the second user; and
determine an interaction history related to the interaction;
an interface coupled to the processor, the interface configured to receive an indication of a preference of the first user for the second user at a second time, the second time being after the first time period, the indication is one of user ratings;
the processor is further configured to:
extract a machine learning feature from the indication, the machine learning feature indicating a compatibility score between the first user and the second user learned from the interaction between the first user and the second user;
update a machine learning pairing algorithm based at least on the indication weighted by the machine learning feature, the machine learning pairing algorithm being configured to generate pairs of users based at least on user profiles and user ratings in response to interactions with other users of the interaction-based recommendation tool; and
determine, by executing the updated machine learning pairing algorithm, to pair the first user with a third user based at least on a second compatibility score between the first user and the third user, that is determined from a first profile of the first user, a second profile of the third user, and the interaction history.
2 . The system of claim 1 , wherein the processor is further configured to, in a first iteration, train the machine learning pairing algorithm to provide a first probability for compatibility between the first user and the second user based at least on a first user profile of the first user and a second user profile of the second user,
wherein updating the machine learning pairing algorithm is performed after the first iteration.
3 . The system of claim 1 , wherein the processor is further configured to dynamically determine when to end the interaction based at least on tracking messages exchanged between the first user and the second user during the interaction, such that as a new message is exchanged between the first user and the second user during the interaction, a duration of the interaction is increased.
4 . The system of claim 1 , wherein the processor is further configured to prevent communication between the first user and the second user at the end of the interaction.
5 . The system of claim 1 , wherein the processor is further configured to assign a weight to the received indication based at least on the interaction,
wherein updating the machine learning pairing algorithm is further based at least on the assigned weight.
6 . The system of claim 5 , wherein the processor is further configured to, in a second iteration, train the machine learning pairing algorithm to provide a second probability for compatibility between the first user and the second user based at least on the interaction history, the assigned weight to the received indication, and the user profiles.
7 . The system of claim 1 , wherein executing the updated machine learning pairing algorithm is in response to the interactions of users of the interaction-based recommendation tool.
8 . The system of claim 1 , wherein the first action comprises the first user requesting to begin the interaction with a user.
9 . The system of claim 1 , wherein the first action comprises the first user indicating availability to begin the interaction with a user.
10 . The system of claim 1 , wherein the first action comprises the first user registering.
11 . The system of claim 1 , wherein the interaction comprises at least one message transmitted from the first user to the second user.
12 . The system of claim 1 , the processor further configured to:
set a timer for a fixed time;
determine that a message is transmitted between the first user and the second user; and
in response to determining that the message is transmitted between the first user and the second user, update the timer.
13 . The system of claim 1 , the processor further configured to:
determine to pair a third user with the first user and the second user; and
pair the first user, the second user, and the third user for the interaction.
14 . The system of claim 1 , wherein determining to pair the first user and the second user is based at least in part on preferences of the first user.
15 . The system of claim 1 , wherein determining to pair the first user and the second user comprises analyzing an interaction history between the first user and the second user.
16 . The system of claim 1 , wherein determining to pair the first user and the second user comprises analyzing an interaction history between the first user and one or more features of the system.
17 . The system of claim 1 , wherein:
the interface is further configured to:
transmit a first prompt to the first user and the second user;
receive a first response from the first user in response to the first prompt;
receive a second response from the second user in response to the first prompt; and
the processor is further configured to in response to receiving the first response from the first user and the second response from the second user, determine to pair the first user and the second user; and
the interface is further configured to display the first response and the second response to the first user and the second user.
18 . A method comprising:
determining, by a processor, that a first user takes a first action with respect to an interaction-based recommendation tool;
extracting, by the processor, a first set of features from a preference history of the first user, the first set of features indicating preferences of the first user with respect to other users;
extracting, by the processor, a second set of features from a preference history of a second user, the second set of features indicating preferences of the second user with respect to other users;
determining, by the processor, to pair the first user and the second user based at least on the first action, the first set of features, and the second set of features;
pairing, by the processor, the first user and the second user for an interaction, the interaction lasting for a first time period, the interaction is concurrent for the first user and the second user;
determining an interaction history related to the interaction;
receiving at a second time, by an interface, an indication of a preference of the first user for the second user, the second time being after the first time period, the indication is one of user ratings;
extracting a machine learning feature from the indication, the machine learning feature indicating a compatibility score between the first user and the second user learned from the interaction between the first user and the second user;
updating, by the processor, a machine learning pairing algorithm based at least on the indication weighted by the machine learning feature, the machine learning pairing algorithm being configured to generate pairs of users based at least on user profiles and user ratings in response to interactions with other users of the interaction-based recommendation tool; and
determining, by executing the updated machine learning pairing algorithm, to pair the first user with a third user based at least on a second compatibility score between the first user and the third user, that is determined from a first profile of the first user, a second profile of the third user, and the interaction history.
19 . The method of claim 18 , further comprising, in a first iteration, training the machine learning pairing algorithm to provide a first probability for compatibility between the first user and the second user based at least on a first user profile of the first user and a second user profile of the second user,
wherein updating the machine learning pairing algorithm is performed after the first iteration.
20 . The method of claim 18 , further comprising dynamically determining when to end the interaction based at least on tracking messages exchanged between the first user and the second user during the interaction, such that as a new message is exchanged between the first user and the second user during the interaction, a duration of the interaction is increased.
21 . The method of claim 18 , further comprising preventing communication between the first user and the second user at the end of the interaction.
22 . The method of claim 18 , further comprising assigning a weight to the received indication based at least on the interaction,
wherein updating the machine learning pairing algorithm is further based at least on the assigned weight.
23 . The method of claim 18 , further comprising in a second iteration, training the machine learning pairing algorithm to provide a second probability for compatibility between the first user and the second user based at least on the interaction history, the assigned weight to the received indication, and the user profiles.
24 . The method of claim 18 , wherein executing the updated machine learning pairing algorithm is in response to the interactions of users of the interaction-based recommendation tool.
25 . The method of claim 18 , wherein the first action comprises the first user requesting to begin an interaction with a user.
26 . The method of claim 18 , wherein the first action comprises the first user indicating availability to begin the interaction with a user.
27 . The method of claim 18 , wherein the first action comprises the first user registering.
28 . The method of claim 18 , wherein the interaction comprises at least one message transmitted from the first user to the second user.
29 . The method of claim 18 , further comprising:
setting, by the processor, a timer for a fixed time;
determining, by the processor, that a message is transmitted between the first user and the second user; and
in response to determining that the message is transmitted between the first user and the second user, updating, by the processor, the timer.
30 . The method of claim 18 , further comprising:
determining, by the processor, to pair a third user with the first user and the second user; and
pairing, by the processor, the first user, the second user, and the third user for the interaction.
31 . The method of claim 18 , wherein determining to pair the first user and the second user is based at least in part on preferences of the first user.
32 . The method of claim 18 , wherein determining to pair the first user and the second user comprises analyzing an interaction history between the first user and the second user.
33 . The method of claim 18 , wherein determining to pair the first user and the second user comprises analyzing an interaction history between the first user and one or more features of a system.
34 . The method of claim 18 , further comprising:
transmitting, by the interface, a first prompt to the first user and the second user;
receiving, by the interface, a first response from the first user in response to the first prompt;
receiving, by the interface, a second response from the second user in response to the first prompt;
in response to receiving the first response from the first user and the second response from the second user, determining, by the processor, to pair the first user and the second user; and
displaying, by the interface, the first response and the second response to the first user and the second user.
35 . At least one non-transitory computer-readable medium comprising a plurality of instructions that, when executed by at least one processor, are configured to:
determine that a first user takes a first action with respect to an interaction-based recommendation tool;
extract a first set of features from a preference history of the first user, the first set of features indicating preferences of the first user with respect to other users;
extract a second set of features from a preference history of a second user, the second set of features indicating preferences of the second user with respect to other users;
determine to pair the first user and the second user based at least on the first action, the first set of features, and the second set of features;
pair the first user and the second user for an interaction, the interaction lasting for a first time period, the interaction is concurrent for the first user and the second user;
determine an interaction history related to the interaction;
receive, at a second time, an indication of preference of the first user for the second user, the second time being after the first time period, the indication is one of the user ratings;
extract a machine learning feature from the indication, the machine learning feature indicating a compatibility score between the first user and the second user learned from the interaction between the first user and the second user;
update a machine learning pairing algorithm based at least on the indication weighted by the machine learning feature, the machine learning pairing algorithm being configured to generate pairs of users based at least on user profiles and user ratings in response to interactions with other users of the interaction-based recommendation tool; and
determine, by executing the updated machine learning pairing algorithm, to pair the first user with a third user based at least on a second compatibility score between the first user and the third user, that is determined from a first profile of the first user, a second profile of the third user, and the interaction history.
36 . The at least one non-transitory computer-readable medium of claim 35 , wherein the instructions are further configured to, in a first iteration, train the machine learning pairing algorithm to provide a first probability for compatibility between the first user and the second user based at least on a first user profile of the first user and a second user profile of the second user,
wherein updating the machine learning pairing algorithm is performed after the first iteration.
37 . The at least one non-transitory computer-readable medium of claim 35 , wherein the instructions are further configured to dynamically determine when to end the interaction based at least on tracking messages exchanged between the first user and the second user during the interaction, such that as a new message is exchanged between the first user and the second user during the interaction, a duration of the interaction is increased.
38 . The at least one non-transitory computer-readable medium of claim 35 , wherein the instructions are further configured to prevent communication between the first user and the second user at the end of the interaction.
39 . The at least one non-transitory computer-readable medium of claim 35 , wherein the instructions are further configured to assign a weight to the received indication based at least on the interaction,
wherein updating the machine learning pairing algorithm is further based at least on the assigned weight.
40 . The at least one non-transitory computer-readable medium of claim 35 , wherein the instructions are further configured to, in a second iteration, train the machine learning pairing algorithm to provide a second probability for compatibility between the first user and the second user based at least on the interaction history, the assigned weight to the received indication, and the user profiles.
41 . The at least one non-transitory computer-readable medium of claim 35 , wherein executing the updated machine learning pairing algorithm is in response to the interactions of users of the interaction-based recommendation tool.
42 . The at least one non-transitory computer-readable medium of claim 35 , wherein the first action comprises the first user requesting to begin the interaction with a user.
43 . The at least one non-transitory computer-readable medium of claim 35 , wherein the first action comprises the first user indicating availability to begin the interaction with a user.
44 . The at least one non-transitory computer-readable medium of claim 35 , wherein the first action comprises the first user registering.
45 . The at least one non-transitory computer-readable medium of claim 35 , wherein the interaction comprises at least one message transmitted from the first user to the second user.
46 . The at least one non-transitory computer-readable medium of claim 35 , wherein the instructions are further configured to:
set a timer for a fixed time;
determine that a message is transmitted between the first user and the second user; and
in response to determining that the message is transmitted between the first user and the second user, update the timer.
47 . The at least one non-transitory computer-readable medium of claim 35 , wherein the instructions are further configured to:
determine to pair a third user with the first user and the second user; and
pair the first user, the second user, and the third user for the interaction.
48 . The at least one non-transitory computer-readable medium of claim 35 , wherein determining to pair the first user and the second user is based at least in part on preferences of the first user.
49 . The at least one non-transitory computer-readable medium of claim 35 , wherein determining to pair the first user and the second user comprises analyzing an interaction history between the first user and the second user.
50 . The at least one non-transitory computer-readable medium of claim 35 , wherein determining to pair the first user and the second user comprises analyzing an interaction history between the first user and one or more features of a system.
51 . The at least one non-transitory computer-readable medium of claim 35 , wherein the instructions are further configured to:
transmit a first prompt to the first user and the second user;
receive a first response from the first user in response to the first prompt;
receive a second response from the second user in response to the first prompt;
in response to receiving the first response from the first user and the second response from the second user, determine to pair the first user and the second user; and
display the first response and the second response to the first user and the second user.