Performing an action based on user interaction data

A non-transitory computer-readable medium storing instructions, the instructions comprising:

• one or more instructions that, when executed by one or more processors of a device,
• cause the one or more processors to:
• obtain historical interaction data concerning a plurality of user interactions;
• obtain historical response data concerning a plurality of user responses associated with the plurality of user interactions;
• process the historical interaction data and the historical response data using a modelling pipeline to determine, for a user interaction of the plurality of user interactions:
• a user interaction score,
• at least one touchpoint set, and
• an association between the user interaction score and the at least one touchpoint set,
• each touchpoint set, of the at least one touchpoint set, including one or more of:
• first touchpoint data indicating who initiated the user interaction,
• second touchpoint data indicating how the user interaction was initiated,
• third touchpoint data indicating why the user interaction was initiated,
• fourth touchpoint data indicating information communicated during the user interaction,
• fifth touchpoint data indicating how long the user interaction lasted,
• sixth touchpoint data indicating a wait time associated with the user interaction,
• seventh touchpoint data indicating whether the user interaction is associated with an existing user interaction issue,
• eighth touchpoint data indicating whether the user interaction was associated with user dissatisfaction, or
• ninth touchpoint data indicating a corrective action taken regarding the user interaction;
• train, based on processing the historical interaction data, a first machine learning model,
• the first machine learning model being trained to determine associations between user interaction scores and touchpoint sets;
• obtain information concerning a current user interaction of a user;
• process the information concerning the current user interaction using the first machine learning model to determine a current user interaction score and a ranking of a plurality of current touchpoint sets for the current user interaction,
• the current user interaction score being based on one or more current touchpoint sets of the plurality of current touchpoint sets,
• determine that the current user interaction score satisfies a threshold;
• process, based on determining that the current user interaction score satisfies the threshold and using a second machine learning model, a highest ranking touchpoint set, of the plurality of current touchpoint sets, to identify an additional touchpoint set,
• the additional touchpoint set being different from the highest ranking touchpoint set, and
• the additional touchpoint set being associated with an additional user interaction score that exceeds the current user interaction score;
• determine, based on the additional touchpoint set, at least one action to be performed;
• and
• cause, based on the additional touchpoint set being associated with the additional user interaction score that exceeds the current user interaction score, the at least one action to be performed.

A method, comprising:

• obtaining, by a device, historical interaction data concerning a plurality of user interactions;
• obtaining, by the device, historical response data concerning a plurality of user responses associated with the plurality of user interactions;
• processing, by the device, the historical interaction data and the historical response data to train a first machine learning model to determine associations between user interaction scores and touchpoints sets,
• the touchpoint sets including one or more of:
• first touchpoint data indicating who initiated a user interaction,
• second touchpoint data indicating how the user interaction was initiated,
• third touchpoint data indicating why the user interaction was initiated,
• fourth touchpoint data indicating information communicated during the user interaction,
• fifth touchpoint data indicating how long the user interaction lasted,
• sixth touchpoint data indicating a wait time associated with the user interaction,
• seventh touchpoint data indicating whether the user interaction is associated with an existing user interaction issue,
• eighth touchpoint data indicating whether the user interaction was associated with user dissatisfaction, or
• ninth touchpoint data indicating a corrective action taken regarding the user interaction;
• obtaining, by the device, information concerning a plurality of current user interactions;
• processing, by the device, the information concerning the plurality of current user interactions to determine information concerning a current user interaction of the plurality of current user interactions;
• processing, by the device, the information concerning the current user interaction to determine a current user interaction score using the first machine learning model,
• the current user interaction score being based on a plurality of current touchpoint sets associated with the current user interaction;
• generating, by the device, using the first machine learning model, and based on the information concerning the current user interaction, a ranking of the plurality of current touchpoint sets for the current user interaction;
• processing, by the device and using a second machine learning model, a highest ranking touchpoint set, of the plurality of current touchpoint sets, to identify an additional touchpoint set,
• the additional touchpoint set being different from the highest ranking touchpoint set, and
• the additional touchpoint set being associated with an additional user interaction score that exceeds the current user interaction score;
• determining, by the device and based on the additional touchpoint set, at least one action to be performed;
• and performing, by the device and based on the additional touchpoint set being associated with the additional user interaction score that exceeds the current user interaction score, the at least one action.