Research on Cigarette Market Capacity Forecasting Based on Data Mining Methods

Abstract

With the tobacco industry gradually stepping into the modern cigarette marketing mode of data and informationization, cigarette precision marketing will become the industry's new way of refined marketing management. Among them, precise placement is one of the important contents of cigarette precision marketing. Precision placement is based on the segmentation, quantification, and combination of brand, customer, market, and time, aiming at precisely supplying the brand to the corresponding market segments, and realizing the goal of “finding the market for the brand, finding the brand for the market; finding the brand for the customer, finding the customer for the brand”. Through the segmentation and quantification of the market and customers, this study this paper can more accurately understand the market demand and customer characteristics, to formulate more targeted marketing strategies. Accurate forecasting of cigarette market capacity is of great significance to enterprises and regulators, and the integration of multiple data processing techniques and forecasting models can help improve the accuracy and applicability of predicting. Specifically, this study aims to construct an accurate customer placement model based on multiple linear regression. After data analysis and model training, the prediction results of the model are validated and evaluated. The results show that the relative error between the model-predicted placement and the actual placement is 1.95%, and the model has high prediction accuracy and stability, which provides valuable reference information for the work of precise cigarette customer placement.