Back to results
US 10,832,218 B1Granted
User interface for visualization of an attrition value
AssigneePalantir Technologies Inc.
InventorsRichard Hsu, Brendan Weickert, Krishnan Aiyer
Life of this patent
Estimating the term…
Abstract
Aspects of the present disclosure involve approaches for calculating an attrition value of a population on the probability that attrition will occur within a predefined time period, and additionally, to calculate an individual attrition value of various features of the population based on their contribution to the calculated attrition value of the population. A computer, such as a server, may receive a cohort definition that includes one or more features and corresponding feature values from a third-party data source or client device. A cohort definition defines a subset of users from among the population of users, based on various combinations of features and feature values.
Claims20
A system comprising:
- one or more processors;
- one or more memories;
- and
- a set of instructions incorporated into the one or more memories, the set of instructions configuring the one or more processors to perform operations for using a network-based attrition-value system to recalculate a visualization of an attrition value of a first subset of users with and using a visualization module to overlay a visualization of an attrition value of a second subset of users over the visualization of the attrition value of the first subset of users in a graphical user interface presented on a client device, the operations comprising:
- based on a receiving of first user input via the visualization module, the first user input identifying a first cohort definition, the first cohort definition including at least a first feature and a feature value of the first feature, the first feature including a user attribute of the first subset of users, identifying the first subset of users from a population of users based on the first cohort definition, using the attrition value system to compute the attrition value of the first subset users based on employment data of the first subset of users using a non-linear-regression statistical technique, and using the visualization module to cause displaying the visualization of the attrition value of the first subset of users within the graphical user interface;
- based on a receiving of a second user input via the visualization module, the second user input identifying a second cohort definition, the second cohort definition including at least a second feature and a feature value of the second feature, the second feature including a user attribute of the second subset of users, using the attrition value system to identify the second subset of users from the population of users based on the second cohort definition and compute the attrition value of the second subset users based on employment data of the second subset of users using the non-linear-regression statistical technique, and using the visualization module to cause the overlaying of the visualization of the attrition value of the second set of users over the visualization of the attrition value of the second set of users within the graphical user interface;
- and
- using the attrition value system to recalculate the attrition value and perform the recalculation of the visualization of the attrition value of the first subset of users based on a detection of an activation, via the visualization module, of a slider element of the user interface, the slider element defining a range of a value included in the employment data.
5 more elements
A method comprising:
- based on a receiving of first user input via a visualization module, the first user input identifying a first cohort definition, the first cohort definition including at least a first feature and a feature value of the first feature, the first feature including a user attribute of the first subset of users, identifying the first subset of users from a population of users based on the first cohort definition, computing, using an attrition value system, the attrition value of the first subset of users using a non-linear-regression statistical technique, and using the visualization module to cause displaying the visualization of the attrition value of the first subset of users within the graphical user interface;
- based on a receiving of a second user input via the visualization module, the second user input identifying a second cohort definition, the second cohort definition including at least a second feature and a feature value of the second feature, the second feature including a user attribute of the second subset of users, using the attrition value system to identify the second subset of users from the population of users based on the second cohort definition and compute the attrition value of the second subset users based on employment data of the second subset of users using the non-linear-regression statistical technique, and using the visualization module to cause overlaying of a visualization of the attrition value of the second set of users over the visualization of the attrition value of the second set of users within the graphical user interface at the client device;
- and
- using the attrition value system to recalculate the attrition value and recalculate the visualization of the attrition value of the first subset of users based on a detection, via the visualization module, of an activation of a slider element of the user interface, the slider element defining a range of a value included in the employment data.
1 more element
A non-transitory machine readable storage medium including instructions that, when executed by one or more process of a machine, cause the machine to perform operations comprising:
- based on a receiving of first user input via a visualization module, the first user input identifying a first cohort definition, the first cohort definition including at least a first feature and a feature value of the first feature, the first feature including a user attribute of the first subset of users, identifying the first subset of users from a population of users based on the first cohort definition, computing, using an attrition value system, the attrition value of the first subset of users using a non-linear-regression statistical technique, and using the visualization module to cause displaying the visualization of the attrition value of the first subset of users within the graphical user interface;
- based on a receiving of a second user input via the visualization module, the second user input identifying a second cohort definition, the second cohort definition including at least a second feature and a feature value of the second feature, the second feature including a user attribute of the second subset of users, using the attrition value system to identify the second subset of users from the population of users based on the second cohort definition and compute the attrition value of the second subset users based on employment data of the second subset of users using the non-linear-regression statistical technique, and using the visualization module to cause overlaying of a visualization of the attrition value of the second set of users over the visualization of the attrition value of the second set of users within the graphical user interface at the client device;
- and
- using the attrition value system to recalculate the attrition value and recalculate the visualization of the attrition value of the first subset of users based on a detection, via the visualization module, of an activation of a slider element of the user interface, the slider element defining a range of a value included in the employment data.
1 more element
Dependencies are inferred from the claim text.