US Patent Application 18323339. LARGE SCALE FORECASTING WITH EXPLANATION INFORMATION FOR TIME SERIES DATASETS simplified abstract

From WikiPatents
Jump to navigation Jump to search

LARGE SCALE FORECASTING WITH EXPLANATION INFORMATION FOR TIME SERIES DATASETS

Organization Name

Oracle International Corporation

Inventor(s)

Chirag Ahuja of Delhi (IN)

Vikas Rakesh Upadhyay of Seattle WA (US)

Samik Raychaudhuri of Bangalore (IN)

Syed Fahad Allam Shah of Washougal WA (US)

Hariharan Balasubramanian of Redmond WA (US)

LARGE SCALE FORECASTING WITH EXPLANATION INFORMATION FOR TIME SERIES DATASETS - A simplified explanation of the abstract

This abstract first appeared for US patent application 18323339 titled 'LARGE SCALE FORECASTING WITH EXPLANATION INFORMATION FOR TIME SERIES DATASETS

Simplified Explanation

- The patent application describes a time series forecasting system. - The system receives a request for a forecast at a specific time point. - The request includes a primary time series dataset and a set of related features. - The system uses a model to generate the forecast. - The model calculates feature importance scores and selects a subset of features. - Attention scores are determined for data points in the primary time series dataset based on the selected features. - The system predicts the actual forecast for the requested time point. - The system also provides explanation information associated with the forecast.


Original Abstract Submitted

A time series forecasting system is disclosed that obtains a time series forecast request requesting a forecast for a particular time point. The forecast request identifies a primary time series dataset for generating the requested forecast and a set of features related to the primary time series dataset. The system provides the primary time series dataset and the set of features to a model to be used for generating the forecast. The model computes a feature importance score for one or more features and selects a subset of features based on their feature importance scores. The model determines attention scores for a set of data points in the primary time series dataset based on the selected subset of features. The system predicts an actual forecast for the particular time point based on the attention scores and outputs the actual forecast and explanation information associated with the actual forecast.