Definição de promoções em supermercado utilizando técnicas de data mining

Abstract

Public and private institutions, including companies, generate a very large and growing volume of data from their business operations. These data result from the activities performed by the institutions that are stored in their databases. It is increasingly evident and verified that these data could be used to support management, set trends and provide information that may be useful and strategic for conducting business. The Data Mining technique is seen as an alternative to extract knowledge from large volumes of data, discovering relationships and patterns, and generating rules to predict and correlate data. This knowledge can assist institutions to make more effective business decisions. In segments such as supermarkets, the quantity of perishables products and the high turnover of products, the agility in terms of deciding which products to buy, the quantity of each, marketing strategies, sales value and others can represent the difference between sale with profit and unnecessary disposal of product, causing financial loss. The data mining allows the companies to identify knowledge about customers’ consumption habits. Aiming to aid the discovery of knowledge in data stored in supermarket databases, an algorithm was implemented using Delphi language and neural network concepts for mining these data. The data obtained from the database mining can be used to help managers and decision-makers.