Machine-learning models for generating emerging user segments based on attributes of digital-survey respondents and target outcomes

A system comprising:

• at least one processor;
• and
• at least one non-transitory computer-readable storage medium storing instructions that, when executed by the at least one processor, cause the system to:
• identify a target outcome associated with survey responses to one or more digital surveys and respondent attributes associated with respondents;
• determine, from the respondent attributes associated with the respondents, a set of deterministic attributes that satisfy a deterministic threshold indicating a threshold measure of confidence for attributes to be deterministic of the target outcome by using a machine-learning model that includes a gradient-based decision tree to determine which of the respondent attributes satisfy the deterministic threshold;
• select a subset of survey responses provided by respondent devices of a subset of the respondents reflecting the set of deterministic attributes;
• generate, based on determining the set of deterministic attributes that satisfy the deterministic threshold and utilizing the machine-learning model to process the subset of survey responses and the set of deterministic attributes, an emerging user segment based on the target outcome by:
• identifying, using the machine-learning model to predict outcomes from the respondent attributes, a set of respondent users predicted to accomplish the target outcome;
• and
• determining a set of nonrespondent users associated with the set of deterministic attributes predicted by the machine-learning model to accomplish the target outcome by:
• identifying a subset of the respondent attributes corresponding to the target outcome based on the subset of survey responses;
• identifying the set of nonrespondent users by utilizing the machine-learning model to determine a semantic similarity between attributes of the set of nonrespondent users and the subset of the respondent attributes;
• and
• grouping the set of nonrespondent users within the emerging user segment based on determining the semantic similarity between attributes of the set of nonrespondent users and the subset of the respondent attributes;
• provide, for display within a graphical user interface of a client device, a segment visualization of the emerging user segment including the set of respondent users and the set of nonrespondent users and a segment-options menu element selectable to modify the segment visualization according to a composition of the emerging user segment;
• and
• modify the segment visualization based on receiving an interaction with the segment-options menu element from the client device.

A method comprising:

• identifying a target outcome associated with survey responses to one or more digital surveys and respondent attributes associated with respondents;
• determine, from the respondent attributes associated with the respondents, a set of deterministic attributes that satisfy a deterministic threshold indicating a threshold measure of confidence for attributes to be deterministic of the target outcome by using a machine-learning model that includes a gradient-based decision tree to determine which of the respondent attributes satisfy the deterministic threshold;
• selecting a subset of survey responses provided by respondent devices of a subset of the respondents reflecting the set of deterministic attributes;
• generating, based on determining the set of deterministic attributes that satisfy the deterministic threshold and utilizing the machine-learning model to process the subset of survey responses and the set of deterministic attributes, an emerging user segment based on the target outcome by:
• identifying, using the machine-learning model to predict outcomes from the respondent attributes, a set of respondent users predicted to accomplish the target outcome;
• and
• determining a set of nonrespondent users associated with the set of deterministic attributes predicted by the machine-learning model to accomplish the target outcome by:
• identifying a subset of the respondent attributes corresponding to the target outcome based on the subset of survey responses;
• identifying the set of nonrespondent users by utilizing the machine-learning model to determine a semantic similarity between attributes of the set of nonrespondent users and the subset of the respondent attributes;
• and
• grouping the set of nonrespondent users within the emerging user segment based on determining the semantic similarity between attributes of the set of nonrespondent users and the subset of the respondent attributes;
• providing, for display within a graphical user interface of a client device, a segment visualization of the emerging user segment including the set of respondent users and the set of nonrespondent users and a segment-options menu element selectable to modify the segment visualization according to a composition of the emerging user segment;
• and
• modifying the segment visualization based on receiving an interaction with the segment-options menu element from the client device.