DESIGN OF METADATA MANAGEMENT PLATFORM USING ARTIFICIAL INTELLIGENCE

The growth rate of data is increasing rapidly every passing day. In addition to this structurally growing data, unstructured data is now also a part of the data world. Today, while many different types of devices produce and transfer data, data is now an asset and value for institutions. However, at a point where data grows and diversifies so rapidly, managing the data itself and the metadata containing the data of the data, benefiting from this, and ensuring data-driven business transformations are even more difficult areas. In this study, a system is presented to companies to solve management problems in the field of metadata, where they can track data and digital assets that are rapidly expanding with the age of digitalization end-to-end. In addition, this system aims to group data with the support of large language models, classify data baskets with machine learning methods, comply with data security policies required by KVKK with natural language processing methods, and create a platform where companies can analyze their own metadata. With this platform, the design phase of which has been completed, using machine learning methods including natural language processing and quality assessment methods, data profiling, increasing data quality, and grouping related data will enable institutions to use the full potential of their data in decision-making. In addition, institutions will be able to manage data lines on the same platform without the need for other tools.