ANN PostgreSQL extension

Abstract

This bachelor thesis deals with the creation of the PostgreSQL extension, which allow us to perform an approximate search for the K nearest neighbours. The search is performed using SQL queries over the data in the database. The goal is to create a PostgreSQL extension with a new data type, which we define in the database and will be used to store vector data. Above this type, the operation of searching for the K nearest neighbours by brute force is further defined. Subsequently, also using the library for the approximate search of the K nearest neighbours. Out of the stated goals, the creation of a new data type for storing vector data and searching for the K nearest neighbours by brute force was achieved.