Back to results
US 12,613,875 B2Granted
Systems and methods for maintaining sequential relationships in aggregating time-series data streams using time shifted event triggers
AssigneeCitibank, N.A.
InventorsThomas Francis Gianelle, Ernst Wilhelm Spannhake, II, Milan Shah
Life of this patent
Estimating the term…
Abstract
Systems and methods are described herein for novel uses and/or improvements to data aggregation related to artificial intelligence applications, specifically applications related to aggregating time-series data. As one example, systems and methods are described herein for predicting effects when aggregating time-series data and modifying the one or more data streams used to populate a model profile and/or feed an artificial intelligence application with the time-series data.
Claims20
A system for maintaining sequential relationships in aggregated time-series data streams using time shifted event triggers, the system comprising:
- one or more processors;
- and
- a non-transitory computer-readable medium comprising instructions that when executed by the one or more processors cause operations comprising:
- receiving a request, via a user interface, to perform a data aggregation procedure based on the aggregated time-series data streams;
- receiving a first data set, wherein the first data set comprises first time-series data beginning at a first time point, wherein the first data set comprises a first initial state characteristic, and wherein the first data set comprises non-synthetic data;
- generating a synthetic historic data set comprising aggregated time series data by:
- receiving, via the user interface, a first required future state characteristic for the first data set;
- retrieving a plurality of historic data sets, wherein each of the plurality of historic data sets comprises respective historic time-series data beginning at a respective time point before the first time point;
- filtering the plurality of historic data sets to determine a first subset of historic data sets based on comparing a first characteristic in each historic data set of the plurality of historic data sets during a first time range to the first initial state characteristic and comparing a second characteristic in each historic data set in the plurality of historic data sets during a second time range to the first required future state characteristic;
- and
- aggregating the first subset of historic data sets to generate the synthetic historic data set beginning at a second time point;
- determining an amount of time-shifting required by time points in the synthetic historic data set to correspond to time points in the first data set by determining a first difference between the first time point and the second time point;
- determining a time point in the synthetic historic data set to apply the amount of time-shifting to by determining an event-trigger pairing in the synthetic historic data set, wherein the event-trigger pairing comprises a first data point and a second data point, wherein the first data point is a trigger and the second data point is an event, and wherein the trigger is at a third time point;
- based on the trigger being at the third time point, time-shifting the third time point based on the first difference to determine a fourth time point;
- executing the data aggregation procedure for the first data set at the fourth time point, wherein the data aggregation procedure comprises querying one or more external databases, parsing external data streams from the one or more external databases, and extracting data from the external data streams;
- and
- generating for display, on the user interface, a notification based on the data from the external data streams.
14 more elements
A method for maintaining sequential relationships in aggregated time-series data streams using time shifted event triggers, the method comprising:
- receiving a first data set, wherein the first data set comprises first time-series data beginning at a first time point, and wherein the first data set comprises a first initial state characteristic;
- receiving a first required future state characteristic for the first data set;
- generating a synthetic historic data set comprising aggregated time series data, wherein the synthetic historic data set is based on:
- retrieving a plurality of historic data sets, wherein each of the plurality of historic data sets comprises respective historic time-series data beginning at a respective time point before the first time point;
- filtering the plurality of historic data sets to determine a first subset of historic data sets based on comparing a first characteristic in each historic data set in the plurality of historic data sets during a first time range to the first initial state characteristic and comparing a second characteristic in each historic data set in the plurality of historic data sets during a second time range to the first required future state characteristic;
- and
- aggregating the first subset of historic data sets to generate the synthetic historic data set beginning at a second time point;
- determining an amount of time-shifting required by time points in the synthetic historic data set to correspond to time points in the first data set by determining a first difference between the first time point and the second time point;
- determining a time point in the synthetic historic data set to apply the amount of time-shifting to by determining an event-trigger pairing in the synthetic historic data set, wherein the event-trigger pairing comprises a first data point and a second data point, wherein the first data point is a trigger and the second data point is an event, and wherein the trigger is at a third time point;
- based on the trigger being at the third time point, time-shifting the third time point based on the first difference to determine a fourth time point;
- and
- executing a data aggregation procedure for the first data set at the fourth time point.
9 more elements
One or more non-transitory, computer-readable media, comprising instructions that, when executed by one or more processors, cause operations comprising:
- receiving a first data set, wherein the first data set comprises first time-series data beginning at a first time point, a first initial state characteristic, and a first required future state characteristic;
- generating a synthetic historic data set comprising aggregated time series data, wherein the synthetic historic data set is based on:
- retrieving a plurality of historic data sets, wherein each of the plurality of historic data sets comprises respective historic time-series data beginning at a respective time point before the first time point;
- filtering the plurality of historic data sets to determine a first subset of historic data sets based on comparing a first characteristic in each historic data set of the plurality of historic data sets during a first time range to the first initial state characteristic and comparing a second characteristic in each historic data set in the plurality of historic data sets during a second time range to the first required future state characteristic;
- and
- aggregating the first subset of historic data sets to generate the synthetic historic data set beginning at a second time point;
- determining an amount of time-shifting required by time points in the synthetic historic data set to correspond to time points in the first data set by determining a first difference between the first time point and the second time point;
- determining a time point in the synthetic historic data set to apply the amount of time-shifting to by determining an event-trigger pairing in the synthetic historic data set, wherein the event-trigger pairing comprises a first data point and a second data point, wherein the first data point is a trigger and the second data point is an event, and wherein the trigger is at a third time point;
- based on the trigger being at the third time point, time-shifting the third time point based on the first difference to determine a fourth time point;
- and
- executing a data aggregation procedure for the first data set at the fourth time point.
8 more elements
Dependencies are inferred from the claim text.