Sentence examples for Data-driven forecast from inspiring English sources

In this study, we introduce a new and simpler DSI approach, the learning-based, data-driven forecast approach (LDFA), by combining dimension reduction and machine learning techniques to quickly provide accurate forecast results and reliably quantify corresponding uncertainty in the results.

However, the creation of data-driven forecasting models with Big Data proves to be challenging, since most data-driven approaches have not been designed to work on a distributed environment [8].

However, data-driven forecasting models trained using the available Big Data may be a possible solution.

Additionally, a benchmark comparing the time required for the training and application of data-driven forecasting models on a single computer and a computing cluster is presented.

In the present paper three different data mining techniques are used to obtain the various data-driven forecasting models: a multiple linear regression (MLR) [32], a least absolute shrinkage and selection operator (LASSO) [32], and a random forest [33].

Afterwards, a benchmark comparing the training and evaluation of data-driven forecasting models using different amounts of data, as well as R and Spark on a single computer and Spark on a computing cluster is presented.

To achieve these goals, a test scenario is conducted in which the times needed for training and evaluating data-driven forecasting models on a single computer and in a distributed environment are calculated and compared.

These goals are achieved by comparing the required time for training and applying different data-driven forecasting models on a computing cluster (using Spark) and on a single computer (using R and Spark).

In the test scenario, data-driven forecasting models are trained on a single computer or on a computing cluster using the previously described techniques and an amount of training data corresponding either to 1 day (1D), 1 week (1W), 1 month (1M), 6 months (6M), 1 year (1Y), 5 years (5Y), or 10 years (10Y) of the load time series.

We propose a fully data driven forecasting methodology that combines filter and wrapper approaches for feature selection, including automatic feature evaluation, construction and transformation.

Starting from efficient material use, accurate (data-driven) demand forecasts, eliminating overstock, and refraining from operating offline stores.