Research and Judgment Method of Cigarette Consumption Based on Collaborative Filtering Algorithm and Clustering Algorithm

Abstract

The present invention relates to a method for studying and judging cigarette consumption based on a collaborative filtering algorithm and a clustering algorithm, comprising the following steps: 1) sample collection; collecting sample consumers' preferences for each product specification of cigarettes; 2) feature collection: selecting multiple users The eigenvalues ​​of the feature information construct the consumer feature vector to obtain the consumer feature matrix; select the feature values ​​of multiple commodity feature information to construct the cigarette feature vector and obtain the cigarette feature matrix; Cigarettes are clustered; 4) Calculation of preference: the consumer's preference for cigarettes of each brand is obtained through a collaborative filtering algorithm. The present invention uses the K-means++ algorithm to cluster consumers and cigarettes, and then uses the idea of ​​collaborative filtering algorithm to measure the preference of a certain consumer to a certain cigarette, and then predicts the consumption of individual consumers through the preference of sample consumers need.

Description

technical field [0001] The invention relates to the field of tobacco sales, in particular to a method for researching and judging cigarette consumption based on the combination of a collaborative filtering algorithm and a K-means++ clustering algorithm, so as to research and judge the consumption demands of individual consumers. Background technique [0002] Cigarettes have "dual attributes" of both value and harm, which determines that the tobacco industry must implement the national tobacco monopoly system. However, with the development of the economy and society, the market-oriented requirements of the tobacco industry are constantly increasing, and the balance between planning and the market is constantly tilting towards the market. In order to grasp the market demand more realistically, produce more marketable products, and carry out more scientific supply of goods, the National Bureau strategically proposed to carry out market demand forecasting work. Since 2003, the ...

AI Technical Summary

Problems solved by technology

At present, there are few domestic researches on methods of forecasting individual consumer demand for cigarette products

Images