| Popis: |
Sales forecasting is an important aspect of modern market intelligence. A reliable revenue forecast will help a company preserve capital on unnecessary product, prepare better for the future, and increase profit. The estimation of grocery sales is associated with predicting the potential sales of stores such as supermarkets, retail outlets, and bakeries. It enables businesses to effectively distribute capital, forecast realistic sales income, and prepare a stronger strategy for the store’s potential development. Conventional forecasting system fails to compete with big data and the precision of revenue forecasting. These problems may be solved by using different data processing strategies. This paper focuses on product sales predictive analytics challenges centered on historical sell data and time-series analysis. Future sales Pre-diction is done by using a hybrid combination of Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). Proposed model is tested on a real-time bigmart dataset obtained from the local market shops. According to the Predictive approach used, the estimates obtained indicated that the number generated does reflect the actual data where the maximum degree of precision was 97 percent and the minimum degree of precision was only beyond 82 percent. |
