Data Mining Of Weighted Association Rule For The Catering Consumption Data

With the development of China’s economy, people’s living standard improve increasingly, and the catering has also developed rapidly. With the growth of consumption demand and fierce competition, catering entrepreneurs continuous to reform and innovate so that they can adapt with competitive strength.The age of big data is coming, it can find valuable information from business data that will bring more profit to enterprise. Many restaurants use the Restaurant Management System to store data which produced all the year around. We analyze this data to find some information such as consuming habits, characteristics of different consumer groups and dishes sale etc, these information provide the basis for marketing strategies.This thesis use association analysis method applying to the data of a restaurant to find association rules of different dishes. The widespread used Algorithm--Apriori Algorithm has its shortage:all items in a basket database are treated uniformly, while every item has different importance in the real world. For example, merchants always get more interested to the higher profitable products, and ignore the lower profitable ones. We generalize this to the case where items are given weights to reflect their importance to the user, and come up with a new Algorithm. The new Algorithm is efficient, because it not only find important rules, but also keep feature of the Apriori Algorithm.Association analysis always use support-confidence framework to mine data, but it will mine out superabundance patterns, which of them are false. So we measure the mined patterns using chi-square(χ2) for correlation from statistics.