Machine-learning models applied to interaction data for determining interaction goals and facilitating experience-based modifications to interface elements in online environments

A computing system comprising:

• means for identifying interaction data associated with user interactions, by a set of users via a plurality of user computing devices, with a user interface of an interactive computing environment, wherein each of the user interactions comprise a set of click interactions with interface elements of an online platform over one or more sessions within the online platform;
• means for computing goal clusters of the interaction data by performing a clustering operation based on sequences of the click interactions of the set of users with the interface elements, wherein the goal clusters represent interaction goals comprising a goal action or a goal interface display of the set of users performing the sequences of the click interactions;
• means for performing inverse reinforcement learning on the goal clusters to return rewards and policies corresponding to each of the goal clusters;
• means for assigning additional sequences of the user interactions with the user interface to the goal clusters;
• means for computing one or more value differentials between a current click state and a projected click state, wherein the value differential is a difference in a base value of the current click state and the projected click state;
• means for computing interface experience metrics of the additional sequences of the user interactions with the user interface using the one or more value differentials between the current click state and the projected click state, the rewards, and the policies corresponding to the goal clusters of the additional sequences;
• and
• means for transmitting the interface experience metrics to the online platform, wherein the interface experience metrics are usable for changing arrangements of interface elements to improve the interface experience metrics.

A non-transitory computer-readable medium having program code that is stored thereon, the program code executable by one or more processing devices for performing operations comprising:

• identifying interaction data associated with user interactions, by a set of users via a plurality of user computing devices, with a user interface of an interactive computing environment, wherein each of the user interactions comprise a set of click interactions with interface elements of an online platform over one or more sessions within the online platform;
• computing goal clusters of the interaction data by performing a clustering operation based on sequences of the click interactions of the set of users with the interface elements, wherein the goal clusters represent interaction goals comprising a goal action or a goal interface display of a set of users performing the sequences of the click interactions;
• performing inverse reinforcement learning on the goal clusters to return rewards and policies corresponding to each of the goal clusters;
• assigning additional sequences of the user interactions with the user interface to the goal clusters;
• computing one or more value differentials between a current click state and a projected click state, wherein the value differential is a difference in a base value of the current click state and the projected click state;
• computing interface experience metrics of the additional sequences of the user interactions with the user interface using the one or more value differentials between the current click state and the projected click state, the rewards and the policies corresponding to the goal clusters of the additional sequences;
• and
• transmitting the interface experience metrics to the online platform, wherein the interface experience metrics are usable for changing arrangements of interface elements to improve the interface experience metrics.