Abstract:

This article describes the development of a recommender system for selecting players based on machine learning. The system introduced the example of hockey with the possibility of expanding its use in various team sports. For each sport different roles and characteristics of the players were considered. The article analyzes information about hockey, football, basketball and volleyball. The characteristics of the players are structured and divided into general groups. For each parameter coefficients are displayed that show the impact on the result of the match. Various machine learning algorithms were used to build the model. The web interface of the application has been created.

Abstract:

The process of automating the management of the reading operation of related data from relational and non-relational databases is described.

The developed software tool is based on the use of the JPA (Java Persistence API) standard, which defines the capabilities of managing the lifecycle of entities in Java applications. An architecture for embedding in event processes has been designed, allowing the solution to be integrated into projects regardless of which JPA implementation is used. Support for various data loading strategies, types, and relationship parameters has been implemented. The performance of the tool has been evaluated.

Abstract: This article describes the development of a system for detecting duplicate video files based on the method of obtaining and analyzing the color map of a video sequence, as an approach that does not require large computational resources, is applicable to a wide range of types of video clips and has small distortions relatively to the original.

Abstract: An approach to automatically generating questions for tests or quizzes using the DBPedia knowledge graph is presented here. The selected knowledge graph has about 5 million entities. DBpedia SPARQL endpoint the ability to make queries to the semantic network using the SPARQL language. The algorithm, the basic queries to the knowledge graph for constructing questions, a non-standard approach to the search for entities are presented in this article.

Abstract:

The development of an application based on machine learning methods for automatic annotation of web pages according to the Microdata standard is described, with the possibility of extension to other standards and injecting data to JSX files. Datasets were collected and prepared for training Machine Learning (ML) models. The ML model metrics were collected and analyzed.