TY  - JOUR
AU  - Ahmed, Rana Shakeel
AU  - Hasnain, Muhammad
AU  - Mahmood, Muhammad Hamza
AU  - Mehmood, Muhammad Abid
PY  - 2024
DA  - 2024/10/20
TI  - Comparison of Deep Learning Algorithms for Retail Sales Forecasting
JO  - ICCK Transactions on Intelligent Systematics
T2  - ICCK Transactions on Intelligent Systematics
JF  - ICCK Transactions on Intelligent Systematics
VL  - 1
IS  - 3
SP  - 112
EP  - 126
DO  - 10.62762/TIS.2024.300700
UR  - https://www.icck.org/article/abs/TIS.2024.300700
KW  - sales forecasting
KW  - deep learning
KW  - CNN
KW  - LSTM
KW  - retail analytics
AB  - We investigate the use of deep learning models for retail sales forecasting in this research. Proper sales forecasting can lead to optimization in inventory management, marketing strategies, and other core business operations. This research evaluates deep learning models such as Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Multilayer Perceptron (MLP), and a hybrid CNN-LSTM model. The models are further improved by adding dense layers to process daily sales data from a major pharmaceutical company. The models are trained on 80% of the dataset and tested on the remaining 20%. Model performance is compared using Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The results indicate that the CNN-LSTM model outperforms the others, achieving the lowest RMSE and MAE values, making it the most suitable for sales forecasting in this context. This research contributes to the field by demonstrating the superiority of hybrid models in handling complex temporal data for predictive analytics. Future work will explore the integration of additional data sources and advanced deep learning architectures to further improve forecasting accuracy and applicability.
SN  - 3068-5079
PB  - Institute of Central Computation and Knowledge
LA  - English
ER  - 

@article{Ahmed2024Comparison,
  author = {Rana Shakeel Ahmed and Muhammad Hasnain and Muhammad Hamza Mahmood and Muhammad Abid Mehmood},
  title = {Comparison of Deep Learning Algorithms for Retail Sales Forecasting},
  journal = {ICCK Transactions on Intelligent Systematics},
  year = {2024},
  volume = {1},
  number = {3},
  pages = {112-126},
  doi = {10.62762/TIS.2024.300700},
  url = {https://www.icck.org/article/abs/TIS.2024.300700},
  abstract = {We investigate the use of deep learning models for retail sales forecasting in this research. Proper sales forecasting can lead to optimization in inventory management, marketing strategies, and other core business operations. This research evaluates deep learning models such as Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), Multilayer Perceptron (MLP), and a hybrid CNN-LSTM model. The models are further improved by adding dense layers to process daily sales data from a major pharmaceutical company. The models are trained on 80\% of the dataset and tested on the remaining 20\%. Model performance is compared using Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The results indicate that the CNN-LSTM model outperforms the others, achieving the lowest RMSE and MAE values, making it the most suitable for sales forecasting in this context. This research contributes to the field by demonstrating the superiority of hybrid models in handling complex temporal data for predictive analytics. Future work will explore the integration of additional data sources and advanced deep learning architectures to further improve forecasting accuracy and applicability.},
  keywords = {sales forecasting, deep learning, CNN, LSTM, retail analytics},
  issn = {3068-5079},
  publisher = {Institute of Central Computation and Knowledge}
}