Determining customer attrition risk

1. A system for determining customer attrition risk based on memory contents of counterpart regions of parameter spaces from a plurality of equivalently configured tree data structures using machine learning operations performed by an external memory storing computer program instructions which when executed by the at least one processor cause the at least one processor to perform operations comprising:

instantiating, in the memory, a first tree data structure, a second tree data structure, and additional tree data structures, each such data structure being equivalently configured in form;

identifying, in the first tree data structure, a first region of a parameter space based on a spatial location of the first region within the first tree data structure;

determining, based on training data associated with the first region of the first data structure, a first estimated value of customer attrition derived from random forest machine learning operations;

identifying, in the second tree data structure, a counterpart second region of a parameter space based on a spatial location of the second region within the second data structure;

determining, based on training data associated with the second region of the second data structure, a second estimated value of customer attrition derived from random forest machine learning operations;

determining that the first region and the second region occupy the same spatial region of their respective data structures, resulting in their being counterpart parameter space regions;

accessing additional estimated values of customer attrition derived from random forest machine learning operations from additional counterpart parameter space regions of additional equivalently configured tree data structures;

determining a standard deviation value based on the first estimated value of customer attrition, the second estimated value of customer attrition, and additional estimated values or customer attrition values;

determining, by an external API employing machine learning, consumer sentiment scores of communications received from a customer by assigning:

a score above zero for text of a customer communication determined to have a positive sentiment,

a score of zero for text of a customer communication determined to have a neutral sentiment, and

a score below zero for text of a customer communication determined to have a negative sentiment;

storing an average sentiment score obtained by averaging sentiment scores of communications received from the customer in a given time period;

displaying, on a graphical user interface, a first linear visual representation based on a first scenario comprising a first customer parameter on an x-axis representing a percent of customer communications having a negative sentiment, and a second customer parameter on a y-axis representing a percent decrease in volume of sales to the customer over the prior year;

displaying, on the graphical user interface, a second linear visual representation based on a second scenario comprising the customer parameter on the x-axis representing a percent of customer communications having a negative sentiment, and the second customer parameter on the y-axis representing a percent decrease in volume of sales to the customer over the prior year; and

configuring each of the linear visual representations on the display such that by actuating or interacting with the graphical user interface the user can alter a displayed scenario to cause the system to display an additional attrition scenario.