Toll Free Helpline (India): 1800 1234 070
Rest of World: +91-9810852116
Free Publication Certificate
Submit Manuscript
Journal code(s)
- E-ISSN Number: 2277-7695
- P-ISSN Number: 2349-8242
- CODEN Code: PIHNBQ
- ZDB-Number: 2663038-2
- IC Journal ID: 7725
Main Menu
Special Issue
Copyright Form
Identifier
Vol. 8, Special Issue 4 (2019)
Time series forecasting in retail: A comprehensive review of deep learning models for sales prediction
Author(s):
Deepak Dembla, SP Singh and Praveen Kumar
Deepak Dembla, SP Singh and Praveen Kumar
Abstract:
Accurate sales forecasting is the backbone of a successful retail operation, impacting everything from inventory management to marketing, customer service, and financial planning. However, the abundance of digital data challenges traditional forecasting methods, demanding advanced analysis techniques. This paper tackles this challenge by conducting a comprehensive review of deep learning models for sales prediction within the retail industry.
Accurate sales forecasting is the backbone of a successful retail operation, impacting everything from inventory management to marketing, customer service, and financial planning. However, the abundance of digital data challenges traditional forecasting methods, demanding advanced analysis techniques. This paper tackles this challenge by conducting a comprehensive review of deep learning models for sales prediction within the retail industry.
Using the rich Citadel POS dataset (2013-2018), we perform a comparative analysis of various machine learning methods. We implement and evaluate both regression (Linear, Random Forest, Gradient Boost) and time-series models (ARIMA, LSTM) to identify the most effective approach for retail sales forecasting.
Our findings reveal that Xgboost outperforms both time-series and other regression models, achieving the highest accuracy with a Mean Absolute Error (MAE) of 0.516 and Root Mean Squared Error (RMSE) of 0.63. This suggests that Xgboost's ensemble learning approach is particularly suited for capturing complex patterns and relationships within retail sales data.
This paper contributes to the advancement of retail sales forecasting by:
Proposing a comprehensive review of deep learning models for retail sales prediction.
Conducting a rigorous comparative analysis of diverse machine learning techniques on a real-world retail dataset.
Identifying Xgboost as the most effective model for this specific task, highlighting its potential for enhancing retail operations.
The insights gained from this study offer valuable guidance for retailers seeking to leverage the power of machine learning for optimized inventory management, targeted marketing campaigns, and improved financial planning.
Pages: 26-29 | 194 Views 85 Downloads
How to cite this article:
Deepak Dembla, SP Singh and Praveen Kumar. Time series forecasting in retail: A comprehensive review of deep learning models for sales prediction. The Pharma Innovation Journal. 2019; 8(4S): 26-29. DOI: 10.22271/tpi.2019.v8.i4Sa.25259
Journal code(s)
- E-ISSN Number: 2277-7695
- P-ISSN Number: 2349-8242
- CODEN Code: PIHNBQ
- ZDB-Number: 2663038-2
- IC Journal ID: 7725
Main Menu
Special Issue
Copyright Form
Identifier
The Pharma Innovation Journal
