Journal of Physics: Conference Series PAPER • OPEN ACCESS Teaching Effect and Improvement Model of College Basketball Sports Based on Big Data Analysis To cite this article: Yanping Liu 2020 J. Phys.: Conf. Ser. 1533 042056 View the article online for updates and enhancements. This content was downloaded from IP address 185.46.84.139 on 10/09/2020 at 09:48 Content from this work may be used under the terms of theCreative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. Published under licence by IOP Publishing Ltd ICAIIT 2020 Journal of Physics: Conference Series 1533 (2020) 042056 IOP Publishing doi:10.1088/1742-6596/1533/4/042056 1 Teaching Effect and Improvement Model of College Basketball Sports Based on Big Data Analysis Yanping Liu* School of Physical education, Shanxi Normal University, Linfen, Shanxi, 041004, China *Corresponding author e-mail: liuyanping3366@163.com Abstract. Basketball has a large audience in colleges and universities(CAU) in China. College basketball has developed rapidly, and its attention has been increasing. It has promoted the disruptive change of Internet technology and brought unprecedented opportunities and challenges to competitive sports information systems. The purpose of this article is to study the effect and improvement model of college basketball sports teaching based on big data analysis (BDA). On the basis of systematically combing the characteristics, concepts, and development trends of big data, this study analyzes the teaching effects and improvement models of basketball sports in China's CAU in the context of the era. The research results show that 60.8% of CAU in China currently use big data technology in basketball sports, which can promote the development of college basketball games, improve athletes' level of competition, and enhance the enjoyment and interest of basketball games. In addition, 90.6% of college physical education teachers believe that teaching based on BDA can also effectively help the team's daily training, arrange training plans for coaches and give guidance. Keywords: Big Data Analysis, College Basketball Sports, Teaching Effect, Model Improvement. 1. Introduction It is common to use data mining technology, digital image processing technology and other means to process basketball games. According to these large and accurate data, the coach can make it easier for the coach to formulate or adjust the game strategy in a timely manner.It can also evaluate the status of the player more objectively, and can also simulate and predict the combination between players, so that a more suitable combination of players can play . Not only that, it can also provide these data to the media and fans, so that they can keep abreast of the player dynamics of the teams they care about, and provide big data support for propaganda teams. The conclusions obtained through big data technology can also effectively help the team's daily training, arrange training plans for coaches and give guidance, and provide players with an opportunity to understand their sports status. This is the practical application of big data technology in basketball games. As a result, the popularity of basketball among the people will increase. With the popularity, the construction of stadiums and stadiums will be accelerated, the utilization rate of idle stadiums will be increased, and the ICAIIT 2020 Journal of Physics: Conference Series 1533 (2020) 042056 IOP Publishing doi:10.1088/1742-6596/1533/4/042056 2 development of the basketball industry system and sports economy will be more rapid. . Whether it is the daily teaching process or the final learning effect detection, they rely on the teacher for demonstration, and then the students practice and display it.The teacher gives Results, it is difficult for teachers to truly evaluate the effect of students' learning practice [1, 2]. The advantage of big data is that it is more objective and accurate. It can be comprehensively and objectively evaluated on the effect of physical training of students in the comprehensive use of teaching and testing. The amount of data in college basketball matches is huge, and the data form is generally unstructured.Data mining techniques based on BDA can be used in new data sources. Use perspectives and new methods to find meaningful, valuable, and hidden data in basketball games [3, 4]. It provides reference basis and services for decision-making, has targeted the coaching team to formulate training and competition strategies, and has the function of intelligent decision support and scientific guidance for the team to establish its own team building system [5]. In this paper, based on the analysis of "big data", Shanxi college basketball informatization teaching is taken as the research object, and the current situation of Shanxi college basketball informatization teaching is deeply understood. Through investigation and analysis on the basic conditions, teaching content, teaching process, teaching methods, teachers 'and students' attitudes and needs for basketball informatization teaching in Shanxi CAU based on BDA, analyze and find out the CAU in Shanxi Problems existing in the development of basketball informatization teaching. 2. Method 2.1 Big Data and Basketball Big Data 4V characteristics of big data. Volume: The amount of data is large and new data may be generated. The scale of big data is in an increasing process. The scale of a certain industry field's data set ranges from several terabytes to tens of petabytes; Variety: With the development of technology and changes in people's living habits, many new types of multi-structure data have appeared, such as the network and smart terminals. A new processing model is required. Information assets that make data more detailed, comprehensive, and systematic [6-8]. 2.2 Algorithm Research 2.2.1 Dynamic Neighborhood Radius The original CABWAD algorithm, the reason why the variable density distribution of data objects cannot be effectively processed, is mainly because the algorithm uses the global neighborhood radius, that is, the neighborhood radius is required to remain unchanged during the entire clustering process. However, in practical applications, if the distribution density is closely surrounded in the process of processing clustering, and the neighborhood radius is gradually adjusted and controlled, data clusters with different distribution densities can be better clustered. That is to say, in the process of cluster analysis, different data clusters have different domain radii due to different distribution densities [9]. Therefore, in the clustering process of the improved algorithm proposed in this paper, it is necessary to abandon the global neighborhood radius adopted by the original algorithm and use a dynamic neighborhood radius. Specifically, its mathematical description formula is: The dynamic neighborhood radius adaptive density reachable distance is defined as: 