(Measuring The Effect of Free Zones on Foreign Trade with the Data Mining and A Sectorial Application)

Abstract

Purpose – In this study, it is aimed to group the foreign trade activities of the freezones in Turkey by measuring them according to specific variables. The goal of this grouping to create a reference point for freezones to be used more efficiently. Design/methodology/approach – The data used in research consists of TUIK foreign trade data. Analysis was realized according to the data found between the years 2003-2018. Analysis has been done using the SPSS program. In this study, clustering analysis has been used. The separation of free zones into clusters was achieved by the K-Means method. Results – According to the results of the research, determined hypotheses have been provided. According to these hypotheses, the free zones in Turkey differ from according to export and import data. According to this data, free zones constitute clusters by specific product classes. Discussion – According to the results obtained from the research, the free zones are clustered according to which references. According to the results found in the study, two different situations have arisen about free zones. The first option is to ensure that the free zones are specialized according to the clusters created. The other option is that each free zone achieves a certain level of trade volume in each product group.