Prediction Of Vegetable Commodity Pricing Based on Machine Learning Algorithms

Abstract

This paper is dedicated to improving the profit of vegetable category goods in superstores and optimizing how to make efficient pricing strategy for vegetable category so as to maximize the revenue of superstores. Therefore, a pricing strategy model based on genetic algorithm for vegetable category goods is proposed. First, we analyze and calculate the Pearson correlation of sales volume between categories, and obtain the positive correlation of leafy and cauliflower categories and the negative correlation of aquatic root and eggplant categories. Then, the cost, price and profit margin of each vegetable category are calculated in a weighted way, and Spearman's correlation coefficient is applied to reveal the relationship between total sales volume and cost. At the same time, we find out the linear relationship between the total sales volume and the weighted sales price of vegetable commodities, and finally, with the total profit as the objective function and the linear regression equation between the total sales volume and the weighted sales price of vegetable commodities as the constraints, we use genetic algorithms to search for the optimum, find out the sales price and sales volume under the maximal profit, and then realize the formulation of the pricing strategy for vegetable commodities.