A Novel Method for Exploring the Store Sales Forecasting using Fuzzy Pruning LS-SVM Approach

Abstract

Intelligent robots, intelligent mobiles, intelligent stores, and so on are just a few of the areas where computer-aided ergonomics is being put to use. Convenience stores (CVS) are adapting to a new era of competition by offering a wider variety of products and services than ever before, such as daily fresh meals, a cafe, ticketing, and a grocery. Therefore, it is becoming increasingly difficult to estimate daily sales of’ fresh commodities due to the impact of both internal and external factors. In the long run, a trustworthy sales-forecasting system is going to be critical for enhancing corporate plans and gaining an edge over the competition. In today's internet age, data production has reached unprecedented levels, well beyond what any single human being can comprehend. This has led to the development of a plethora of machine learning methods. In this proposed approach various machine learning methods are explored for predicting store's sales and evaluate them to find the one that works best for the specific scenario. Training times are reduced and data quality is enhanced with the help of Normalization in the proposed approach. K-Means is a popular feature selection clustering algorithm. Fuzzy Pruning LS-SVM is used in the suggested method for training the model. The proposed model has superior performance on SVM and CNN.