Automatic computer prediction of resource attrition

One or more non-transitory storage media storing instructions which, when executed by one or more computing devices, cause performance of a method, the method comprising:

• receiving digital datasets representing and storing a plurality of digital features corresponding to each of a plurality of resources,
• the plurality of digital features corresponding to a time point,
• the plurality of resources being associated with one or more business entities at the time point,
• each of the plurality of digital features having at least a level attribute, an interval attribute, and an aggregate attribute,
• the level attribute having one of a plurality of possible values, including corporate or individual,
• the interval attribute having one of a plurality of possible values, including until the end of the year,
• at least one of the plurality of digital features having an interval attribute having a value of until the end of the year;
• receiving an outcome value specifying whether each of the plurality of resources is disassociated from the corresponding business entity within a certain period of time from the corresponding time point;
• creating a digital model that is programmed to generate an indicator related to whether a certain resource associated with a certain business entity is being disassociated from the certain business entity within the certain period of time from a certain time point based on the plurality of digital features corresponding to each of the plurality of resources and the associated outcomes;
• receiving a specific plurality of digital features corresponding to a specific resource different from the plurality of resources and being associated with a specific business entity, from a device associated with the specific business entity;
• executing the digital model using the specific plurality of digital features;
• transmitting a result of the executing to the device associated with specific business entity.

A computer-implemented method of automatically identifying datasets predictive of potential resource attrition, comprising:

• receiving, by a processor, multiple digital items corresponding to a plurality of resources, the multiple digital items covering a time range;
• the plurality of resources being associated with one or more business entities during the time range,
• each of the multiple digital items corresponding to one of the plurality of resources;
• each of the multiple digital items having a level attribute having a value of corporate or resource;
• receiving, by the processor, a plurality of outcomes respectively corresponding to the plurality of resources, each outcome indicating whether the resource is disassociated from the corresponding business entity at an ending time point at or after the end of the time range;
• causing display of an option to add a feature definition;
• causing, for an added feature definition, display of an option to add one or more attributes,
• the one or more attributes including a level attribute and an aggregate attribute,
• causing, for the level attribute, display of a first plurality of values for selection, including individual or corporate;
• causing, for the aggregate attribute, display of a second plurality of values for selection, including number of changes;
• causing display of an option to select a beginning time point as a beginning of a value of the interval attribute;
• generating a plurality of digital features for each of the plurality of resources by applying added feature definitions to the multiple digital items;
• creating, by the processor, a digital model that is programmed to generate an indicator related to whether a certain resource associated with a certain business entity is being disassociated from the certain business entity within a certain period of time from a certain time point based on the plurality of digital features corresponding to each of the plurality of resources and the associated outcomes;
• in response to receiving a specific plurality of digital features corresponding to a specific resource different from the plurality of resources and being associated with a specific business entity, causing executing the digital model using the specific plurality of digital features and causing display of a result of the executing.