Predicting customer lifetime value with unified customer data
AssigneeAmperity, Inc.
InventorsYan Yan, Aria Haghighi, Nicholas Resnick, Andrew Lim
Life of this patent
Estimating the term…
Abstract
Disclosed are techniques for generating features to train a predictive model to predict a customer lifetime value or churn rate. In one embodiment, a method is disclosed comprising receiving a record that includes a plurality of fields and selecting a value associated with a selected field in the plurality of fields. The method then queries a lookup table comprising a mapping of values to aggregated statistics using the value and receives an aggregated statistic based on the querying. Next, the method generates a feature vector by annotating the record with the aggregated statistic. The method uses this feature vector as an input to a predictive model.
Claims15
A method comprising:
- receiving a record, the record including a plurality of fields;
- selecting a value associated with a selected field in the plurality of fields;
- querying a lookup table using the value, the lookup table comprising a mapping of values to aggregated statistics, wherein querying a lookup table includes determining that the lookup table is out of date by determining if a change in size of a unified dataset storing aggregated statistics exceeds a pre-configured size threshold and updating the lookup table in response;
- receiving an aggregated statistic based on the querying;
- generating a feature vector by annotating the record with the aggregated statistic;
- and
- inputting the feature vector to a predictive model.
A non-transitory computer-readable storage medium for tangibly storing computer program instructions capable of being executed by a computer processor, the computer program instructions defining the steps of:
- receiving a record, the record including a plurality of fields;
- selecting a value associated with a selected field in the plurality of fields;
- querying a lookup table using the value, the lookup table comprising a mapping of values to aggregated statistics, wherein querying a lookup table includes determining that the lookup table is out of date by determining if a change in size of a unified dataset storing aggregated statistics exceeds a pre-configured size threshold and updating the lookup table in response;
- receiving an aggregated statistic based on the querying;
- generating a feature vector by annotating the record with the aggregated statistic;
- and
- inputting the feature vector to a predictive model.
An apparatus comprising:
- a processor;
- and
- a storage medium for tangibly storing thereon program logic for execution by the processor, the stored program logic causing the processor to perform the operations of:
- receiving a record, the record including a plurality of fields;
- selecting a value associated with a selected field in the plurality of fields;
- querying a lookup table using the value, the lookup table comprising a mapping of values to aggregated statistics, wherein querying a lookup table includes determining that the lookup table is out of date by determining if a change in size of a unified dataset storing aggregated statistics exceeds a pre-configured size threshold and updating the lookup table in response;
- receiving an aggregated statistic based on the querying;
- generating a feature vector by annotating the record with the aggregated statistic;
- and
- inputting the feature vector to a predictive model.
Dependencies are inferred from the claim text.