Biased ticket offers for actors identified using dynamic assessments of actors' attributes
Techniques herein attempt to provide actors with more flexible and satisfactory experiences regarding obtaining tickets for an event. A learning model may identify attributes indicative of whether a particular actor (e.g., attempting to purchase tickets to an event) possesses a desirable characteristic (e.g., is likely to attend the event). Each actor can then be evaluated to estimate whether she is a good actor (possesses the characteristic). If so, favored opportunities may be made available, such as the opportunity to buy high-demand tickets. An actor may further have the opportunity to hold or reserve tickets for a period time, during which other actors cannot purchase them. A fee for holding or reserving tickets (and/or maintaining the hold or reserve) can be dynamically set based on market factors. Opportunities to modify seat assignments to allow a group of friends to sit together may also be provided.
1. A computer-implemented method comprising:
continuously collecting attribute data associated with a plurality of user devices, the attribute data representing one or more attributes of a user device of the plurality of user devices, each attribute of the one or more attributes associated with a user device of the plurality of user devices being collected during an interaction between the user device and a ticketing system;
generating a dynamic model characterizing one or more groups of user devices from amongst the plurality of user devices, the characterization of each group of the one or more groups of user devices being indicative of whether or not to grant the group of user devices access to an event, the dynamic model being generated by executing one or more machine-learning techniques on the collected attribute data, and the dynamic model being updated as time progresses and additional attribute data is collected;
classifying, by a server, a target group of the one or more groups, the target group being classified based on the existence or absence of a particular attribute within the attribute data of one or more target user devices associated with the target group, wherein:
the target group corresponds to a plurality of robot users, and
attributes relating to interactions with ticket system indicative of robot behavior;
detecting, by the server, altered activity associated with the one or more target user devices associated with the target group, the altered activity causing at least one target user device of the one or more target user devices to be falsely characterized;
in response to the detection, modifying an aspect of the particular attribute, the modification changing the classifying of the target group;
determining whether or not to exclude each user device included in the target group from accessing a ticket to the event; and
in response to a determination to exclude each user device included in the target group, preventing each user device included in the target group from accessing the ticket to the event.
2. The computer-implemented method of claim 1 , wherein determining whether or not to exclude each user device included in target group is based, at least in part, on a characteristic associated with the target group, the characteristic being user defined or defined based on the one or more attributes associated with the user devices included in the target group.
3. The computer-implemented method of claim 1 , wherein the dynamic model indicates a rate of increase or decrease of the particular attribute included in the collected attribute data.
4. The computer-implemented method of claim 3 , wherein in response to determining the rate of increase of the particular attribute, modifying one or more conditions for including a user device in the target group.
5. The computer-implemented method of claim 3 , wherein in response to determining the rate of decrease of the particular attribute, modifying one or more conditions for excluding a user device from the target group.
6. The computer-implemented method of claim 3 , wherein the target group corresponds to a group of one or more robot users.
7. The computer-implemented method of claim 3 , wherein the target group corresponds to a group of one or more human users.
8. A system, comprising:
one or more processors; and
a non-transitory computer-readable storage medium containing instructions which, when executed on the one or more processors, cause the one or more processors to perform operations including:
continuously collecting attribute data associated with a plurality of user devices, the attribute data representing one or more attributes of a user device of the plurality of user devices, each attribute of the one or more attributes associated with a user device of the plurality of user devices being collected during an interaction between the user device and a ticketing system;
generating a dynamic model characterizing one or more groups of user devices from amongst the plurality of user devices, the characterization of each group of the one or more groups of user devices being indicative of whether or not to grant the group of user devices access to an event, the dynamic model being generated by executing one or more machine-learning techniques on the collected attribute data, and the dynamic model being updated as time progresses and additional attribute data is collected;
classifying, by the one or more processors, a target group of the one or more groups, the target group being classified based on the existence or absence of a particular attribute within the attribute data of one or more target user devices associated with the target group, wherein:
the target group corresponds to a plurality of robot users, and
attributes relating to interactions with ticket system indicative of robot behavior;
detecting, by the one or more processors, altered activity associated with the one or more target user devices associated with the target group, the altered activity causing at least one target user device of the one or more target user devices to be falsely characterized;
in response to the detection, modifying an aspect of the particular attribute, the modification changing the classifying of the target group;
determining whether or not to exclude each user device included in the target group from accessing a ticket to the event; and
in response to a determination to exclude each user device included in the target group, preventing each user device included in the target group from accessing the ticket to the event.
9. The system of claim 8 , wherein determining whether or not to exclude each user device included in target group is based, at least in part, on a characteristic associated with the target group, the characteristic being user defined or defined based on the one or more attributes associated with the user devices included in the target group.
10. The system of claim 8 , wherein the dynamic model indicates a rate of increase or decrease of the particular attribute included in the collected attribute data.
11. The system of claim 10 , wherein in response to determining the rate of increase of the particular attribute, modifying one or more conditions for including a user device in the target group.
12. The system of claim 10 , wherein in response to determining the rate of decrease of the particular attribute, modifying one or more conditions for excluding a user device from the target group.
13. The system of claim 10 , wherein the target group corresponds to a group of one or more robot users.
14. The system of claim 10 , wherein the target group corresponds to a group of one or more human users.
15. A computer-program product tangibly embodied in a non-transitory machine-readable storage medium, including instructions configured to cause a processing apparatus to perform operations including:
continuously collecting attribute data associated with a plurality of user devices, the attribute data representing one or more attributes of a user device of the plurality of user devices, each attribute of the one or more attributes associated with a user device of the plurality of user devices being collected during an interaction between the user device and a ticketing system;
generating a dynamic model characterizing one or more groups of user devices from amongst the plurality of user devices, the characterization of each group of the one or more groups of user devices being indicative of whether or not to grant the group of user devices access to an event, the dynamic model being generated by executing one or more machine-learning techniques on the collected attribute data, and the dynamic model being updated as time progresses and additional attribute data is collected;
classifying, by the processing apparatus, a target group of the one or more groups, the target group being classified based on the existence or absence of a particular attribute within the attribute data of one or more target user devices associated with the target group, wherein:
the target group corresponds to a plurality of robot users, and
attributes relating to interactions with ticket system indicative of robot behavior;
detecting, by the processing apparatus, altered activity associated with the one or more target user devices associated with the target group, the altered activity causing at least one target user device of the one or more target user devices to be falsely characterized;
in response to the detection, modifying an aspect of the particular attribute, the modification changing the classifying of the target group;
determining whether or not to exclude each user device included in the target group from accessing a ticket to the event; and
in response to a determination to exclude each user device included in the target group, preventing each user device included in the target group from accessing the ticket to the event.
16. The non-transitory machine-readable storage medium of claim 15 , wherein determining whether or not to exclude each user device included in target group is based, at least in part, on a characteristic associated with the target group, the characteristic being user defined or defined based on the one or more attributes associated with the user devices included in the target group.
17. The non-transitory machine-readable storage medium of claim 15 , wherein the dynamic model indicates a rate of increase or decrease of the particular attribute included in the collected attribute data.
18. The non-transitory machine-readable storage medium of claim 17 , wherein in response to determining the rate of increase of the particular attribute, modifying one or more conditions for including a user device in the target group.
19. The non-transitory machine-readable storage medium of claim 17 , wherein in response to determining the rate of decrease of the particular attribute, modifying one or more conditions for excluding a user device from the target group.
20. The non-transitory machine-readable storage medium of claim 17 , wherein the target group corresponds to a group of one or more robot users.