Backend of an Event Aggregator with Recommendation System

Every IT company, operating in the B2C or C2C markets, has the goal of increasing user sessions. In the case of Yandex Search, this goal is divided into various segments of queries: educational, medical, financial, sports, etc. One of the most popular segments is entertainment. Besides featuring websites of various companies in the search results, Yandex can create its own product-sites that would enhance the quality of search and allow people to more easily and quickly find the entertainment they are looking for. The service "What to Do" - an event aggregator with a recommendation system, is designed to assist people in this regard. On it, companies can post events and venues for which they sell tickets, thereby providing users with aggregated entertainment options in one place, helping to answer the question: "What to do?". The paper describes the components that allow categorizing, merging events, collecting data for the recommendation system, and preparing the final output for the client. The document is structured as follows: 1) Introduction provides information about the name of the program, a brief description of its application area, and the documents on which the development is based. 2) Chapter 1 contains the functional and operational purpose of the program, analysis of competitors, and details about relevant search queries. 3) Chapter 2 describes the architecture of the program, service components, service algorithms, and technologies. 4) Chapter 3 is dedicated to the results of the work. Consists of 42 pages. Contains 11 sources, 24 images, 18 code snippets. (Code not presented in the report is under NDA)