Examination of big data of different sectors with the association rules approach

In these periods when technology is developing rapidly and the competitive environment of retailers is increasing rapidly, computer use is increasing rapidly. Retail companies need to operate in multi-dimensional databases due to the increase in technology and computer usage. There is a need to find solutions to the data confusion between sales. In this thesis study, it is aimed to reach meaningful data analysis by using the market basket analysis method of data mining by considering the databases of different sectors of the retailer companies and to enable the retailer companies to create future based market stock sales forecasts and sales strategies. Thanks to this study, data mining applications, determining the demands of the retailer companies, forming future sales strategies, storing the data in their structure by converting them into meaningful data, etc. It produces various solutions such as. This study uses the association rules analysis, which is the most common method of data mining in the relevant field. Analysis of databases is carried out using Apriori and Fp-Growth algorithms to uncover the correlation between datasets. Using the Weka software platform, which is the software platform mostly used in data mining technologies, Apriori and Fp Growth algorithm is applied. Application results are discussed and interpreted. In the association rules method, the market results in a basket analysis approach. With this study, it is aimed to ensure that different databases within different retail firms contribute to stock product quantities, stock shelf sequences and sales strategies.