Machine Learning Based Approaches for Short Term Sales Forecasting in E-Commerce

Mehmet Ardıl Altuncu*, Muhammed Hamza Tastan, Tuncay Özcan

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

It is very important for companies with high inventory turnover to be able to efficiently carry out sales and raw material purchases in their trade processes. For this reason, it is very important to be able to predict their short-term sales to execute their own plans in the most effective way. In this study, LSTM, SVR and LR models are proposed to predict short-term sales of companies. For this purpose, 6-month data of a retail company operating in B2B was used. First, to get a more effective result in hourly forecasts, the data, which is a 2-dimensional array, was used in such a way that it would be effective in the last 24 h by including the rolling mechanism in the model. Then, LSTM, SVR and LR models were applied using the dataset developed with the rolling mechanism. The results of the analysis show that, although close to each other, the LSTM model captures the patterns better and that the use of this model can be used as a different option in the management of companies’ short-term sales.

Original languageEnglish
Title of host publicationTowards Industry 5.0 - Selected Papers from ISPR 2022
EditorsNuman M. Durakbasa, M. Güneş Gençyılmaz
PublisherSpringer Science and Business Media Deutschland GmbH
Pages16-24
Number of pages9
ISBN (Print)9783031244568
DOIs
Publication statusPublished - 2023
Event22nd International Symposium for Production Research, ISPR 2022 - Antalya, Turkey
Duration: 6 Oct 20228 Oct 2022

Publication series

NameLecture Notes in Mechanical Engineering
ISSN (Print)2195-4356
ISSN (Electronic)2195-4364

Conference

Conference22nd International Symposium for Production Research, ISPR 2022
Country/TerritoryTurkey
CityAntalya
Period6/10/228/10/22

Bibliographical note

Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Keywords

  • E-Commerce
  • Linear regression
  • LSTM
  • Sales forecasting
  • SVR

Fingerprint

Dive into the research topics of 'Machine Learning Based Approaches for Short Term Sales Forecasting in E-Commerce'. Together they form a unique fingerprint.

Cite this