CONTRIBUTION TO SALES FORECASTING BASED ON RECURRENT NEURAL NETWORK IN THE CONTEXT OF A MOROCCAN COMPANY
Domaine:
socioeconomicType de record:
paper
Créateur:
RguMouBen
Éditeur:
Uni
Éditeur:
CCSD
Hôte:
International audience
Forecasting sales is a prerequisite for successful management, this avoids overproduction and therefore costly storage, the objective of this work is to provide a reliable sales forecasting model to help organizations make strategic and operational decisions. Indeed, this work accounts for the profitability of the use of techniques based on artificial intelligence and more precisely the application of recurrent neural networks for a correct prediction of sales. We initially developed the necessary adjustments of this type of algorithm and thereafter we opted for a justified use of the long short term memory (LSTM) techniques of recurrent neural networks for forecasting sales instead of the conventional forecasting methods already used.in this sense, we have proposed a prediction model based on the time series of previous sales which reacts and readjusts the parameters of each sequence.
Similaires
FORECASTING OF LONG-TERM ELECTRICAL LOAD DEMAND USING RECURRENT NEURAL NETWORKTowards audio-based identification of Ethio-Semitic languages using recurrent neural networkBiomedical Event Trigger Identification Using Bidirectional Recurrent Neural Network Based ModelsRecurrent Neural Network-based Base Transceiver Station Power System Failure PredictionDeep Recurrent Neural Network in Machine Translation from English to KichukaGEEZ TO AFAAN OROMOO MACHINE TRANSLATION USING RECURRENT NEURAL NETWORK