| CPC G06F 9/5077 (2013.01) [G06F 9/505 (2013.01); G06F 11/3414 (2013.01); G06F 18/214 (2023.01); G06N 20/20 (2019.01)] | 20 Claims |
|
1. A system for generating dynamic database query responses using ensemble prediction by correlating probability models with non-homogeneous time dependencies to generate time-specific data processing predictions, the system comprising:
one or more processors; and
a non-transitory, computer-readable media comprising instructions that, when executed by the one or more processors, cause operations comprising:
receiving a first portion of a time series data feed for a first machine learning model, wherein the first portion of the time series data feed comprises a first data processing load prediction based on a real-time average of data processing loads from a first plurality of data sources;
receiving a second portion of the time series data feed for a second machine learning model, wherein the second portion of the time series data feed comprises a second data processing load prediction based on a progressively calculated weighted average of data processing loads from a second plurality of data sources;
receiving a third portion of the time series data feed for a third machine learning model, wherein the third portion of the time series data feed comprises a third data processing load prediction based on a compounded average over a rolling time window of data processing loads from a third plurality of data sources;
determining a first feature input for the first machine learning model based on the first portion of the time series data feed, a second feature input for the second machine learning model based on the second portion of the time series data feed, and a third feature input for the third machine learning model based on the third portion of the time series data feed;
inputting the first feature input into the first machine learning model, the second feature input into the second machine learning model, and the third feature input into the third machine learning model to generate a first output from the first machine learning model, a second output from the second machine learning model, and a third output from the third machine learning model;
comparing the first output to the second output and the third output to detect an outlier result in the first output;
in response to detecting the outlier result, inputting the first output, the second output, and the third output into a fourth machine learning model to generate a fourth output;
determining, based on the fourth output, an aggregate measure of correlation for the first machine learning model, the second machine learning model, and the third machine learning model; and
generating a dynamic database query response based on the aggregate measure of correlation.
|