A Comprehensive Analysis of Retail Sales Forecasting Using Machine Learning and Deep Learning Methods

• Published in: 2023 International Conference on Data Science and Network Security (ICDSNS) pp. 1 - 5
• Main Authors: B S, Suresh, Suresh, M.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 28.07.2023
• Subjects: Auto Regressive Integrated Moving Average; Companies; Convolutional Neural Network; Convolutional neural networks; Deep learning; Exponential Smoothing; Holt-Winter; Long Short-Term Memory; Retail Sales Forecasting; Smoothing methods; Statistical analysis; Time series analysis; Time Series Decomposition
• DOI: 10.1109/ICDSNS58469.2023.10245887

Sales forecasting is important in item production, transportation, and supply chain management, and it has been recognized by both academics and practitioners. A large number of sales forecasting methods have been utilized to forecast sales in the retail industry. Retailers often face confusion when predicting product forecasts, as the same product forecasting can yield diverse and conflicting results. Traditional forecasting methods are capable of predicting future sales behavior under the relationships between the variables and not just the behavior of past and future trends. The forecasting of the products is difficult for large business companies because it has some products and it has much of time. In this study, the retail sales forecasting method is utilized to the effective retail sales forecasting method. This method is classified into two types such as time series analysis and forecasting for sales forecasting. In this method, the various methods like exponential smoothing, time series decomposition, statistical methods of ARIMA, Holt-winter seasonal method as well as deep learning methods of Long Short-Term Memory (LSTM) and Convolutional Neural network (CNN) are used for retail sales forecasting. These methods provide better performance, higher accuracy results, less computational time and training costs as demonstrated in this survey.