Information system for predicting the results of matches in the NBA
Students Name: Drevych Liubomyr Orestovych
Qualification Level: magister
Speciality: Information Systems and Technologies
Institute: Institute of Computer Science and Information Technologies
Mode of Study: full
Academic Year: 2021-2022 н.р.
Language of Defence: ukrainian
Abstract: In recent years, sports forecasting has become popular, as it is seen from the mass financial transactions in sports betting. One of the most popular sports in the world, which attracts betting as well as millions of fans around the world, is basketball, especially the National Basketball Association (NBA) of the United States. Machine learning technology is already widely used in all aspects of the game of basketball. For example, with the help of cameras and wearable sensors, computers can accurately obtain sports data and physiological data of athletes during training or competition. Using machine learning technology to analyze this data can not only help coaches create personalized training plans that will improve player play during the season, but also help the coaching staff develop the best strategies for future games and enable them to predict team results in the standings based on this data, all based on machine learning systems. In the 21st century, more and more software is being created that increases the speed and quality of the data analysis and visualization process, but despite the large number of programs created, there are very few specialized systems. Therefore, the relevant scientific and technical task is to research and create software for data analysis and visualization. Research on decision-making on the basketball court has always been a relevant issue in the sports domain. For these open team sports events on the pitches, the level of sports decision-making directly affects the athlete’s ability to compete, their skills and tactics. This requires athletes to be able to capture goals and process them in real time. Therefore, the relevant scientific and technical task is to research and create software for data analysis and visualization using modern programming languages and visualization of large data sets, which will create a system that would meet the high modern requirements of computer system architecture and UI/UX design for user convenience. The object of research is the process of predicting the results of NBA matches. The subject of research is the methods and tools for predicting the results of NBA matches. The practical value of the results obtained is to create a software product that will provide data collection and processing through which the coaching staff will model and forecast for the season. Research methods. The following methods were used to solve the tasks: - analysis and generalization - in the analysis of existing tools for processing large data sets and their visualization as well as in reviewing the available software for creating models of machine learning; - formalization - in substantiating the architecture for the development of a computer system for data analysis and visualization; - design and programming - applied while using the selected programming languages, frameworks and libraries to create software. The aim of the study. The aim of the study is to analyze the methods and tools for creating an effective information system for analyzing and visualizing open data to predict results. The result of the study was the development of an Information System for predicting the results of matches in the NBA. This system was designed to analyze and visualize team results and player statistics based on open data. This system will enable its users to monitor team results and player statistics as well as to view the current forecast for the next season. In order to obtain the desired result, we conducted the research necessary to create the information system. While creating the information system, we investigated and analyzed the scope of its application, conducted market research and analysis of similar systems that exist in the market. We set a number of requirements to the developed system and studied the typical features of the architecture of such platforms to better meet such requirements. The next stage of development was a systematic analysis of the subject area. During the analysis of the subject area, the creation of this information system was substantiated and the purpose of the system development, its purpose, place of application were determined. The effects after the implementation of this system, functional and non-functional requirements for the system were also determined and the Gantt chart was developed to determine the work schedule. After the analysis of the subject area, a conceptual model of the system was created, which included diagrams of variants, uses, classes, sequences, diagrams of transitions of states and activities, and components. At this stage, the main entities that will be necessary to achieve this goal were identified. For development of this information system methods and means of realization of the task were defined and analyzed. The main components, frameworks and technologies that are available to solve the problem based on the needs of the system have been identified. After analyzing the methods and tools that will be used in the development of the information system, there was a practical part of creating an information system for forecasting results in the NBA. Keywords: basketball, NBA, machine learning, data analysis, UI / UX References: 1) Sampaio, J.; McGarry, T.; Calleja-Gonzalez, J.; Saiz, S.J.; i del Alcazar, X.S.; Balciunas, M. Exploring Game Performance in the National Basketball Association Using Player Tracking Data. ,2015, 10 [Електроний ресурс]: [Веб-сайт]. -Електроні дані. -Режим доступу: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0132894 2) Bourbousson, J.; Seve, C.; McGarry, T. Space–time coordination dynamics in basketball: Part 1. Intra- and inter-couplings among player dyads. J. Sports Sci.2010,28,339–347URL: https://www.tandfonline.com/doi/abs/10.1080/02640410903503632 3) Lamas, L.; Junior, D.D.R.; Santana, F.; Rostaiser, E.; Negretti, L.; Ugrinowitsch, C. Space creation dynamics in basketball offence: Validation and evaluation of elite teams. Int. J. Perform. Anal. Sport 2011 , 11, 71–84. URL: https://www.tandfonline.com/doi/abs/10.1080/24748668.2011.11868530 4) Santana, F.L.; Rostaiser, E.; Sherzer, E.; Ugrinowitsch, C.; Barrera, J.; Lamas, L.; Santana, F.L.; Rostaiser, E.; Sherzer, E.; Ugrinowitsch, C.; et al. Space protection dynamics in basketball: Validation and application to the evaluation of offense-defense patterns. Mot. Rev. Educ. Fisica 2015, 21, 34–44.URL: https://www.scielo.br/j/motriz/a/44n8v3yvMYt7wXDVrcRvpfm/?lang=en 5) Courel-Ibanez, J.; McRobert, A.P.; Toro, E.O.; Velez, D.C. Inside game effectiveness in NBA basketball: Analysis of collective interactions. Kinesiology 2018, 50, 218–227. URL: https://dx.doi.org/10.26582/k.50.2.5