摘要

NBA正在以前所未有的影响力吸引着全世界球迷，每年的NBA总决赛都有上百个国家直播，其在篮球本身，篮球文化，篮球经济方面带来的影响不可估量.篮球比赛最本质的理念就是获得胜利，因此本系统依次从:主场优势，比赛强度和比赛数据三个不同的角度对影响NBA球队比赛结果的因素进行分析。随着人工智能的发展，Python网络爬虫技术应用越来越广。Python是一种多范式的编程语言，既适合面向对象的应用开发，也适合函数式设计模式.Python已然成为数据科学家们在数据分析，可视化和机器学习语言，它可以带来高效率和高生产力。

本文利用Python网络爬虫技术进行NBA球队数据的爬取，主要是采用Scarpy进行网络爬虫，利用pandas， numpy，sklearn等模块对爬取的球队的篮板，抢断等数据进行分析筛选以及可视化处理，用聚类算法数据分析并将数据存储为csv格式文件，同时也探讨了数据挖掘在最佳球员归属，薪资匹配，球员搭档以及比赛预测等应用问题。

关键词：NBA；数据分析；比赛预测；可视化；Python

Abstract

NBA is attracting fans all over the world with an unprecedented influence. Every year, the NBA Finals are broadcast live in hundreds of countries, which has an immeasurable impact on basketball itself, basketball culture and basketball economy. The most essential concept of basketball game is to win, so this system analyzes the factors that affect the results of NBA teams from three different perspectives: home field advantage, game intensity and game data. With the development of artificial intelligence, Python web crawler technology is more and more widely used. Python is a multiparadigm programming language suited to both object-oriented application development and functional design patterns. Python has become the language for data scientists in data analysis, visualization, and machine learning, which can lead to high efficiency and productivity.

This paper uses Python web crawler technology to crawl NBA team data, mainly using Scarpy for web crawler, using pandas, numpy,sklearn and other modules to analyze and filter the crawling team rebounds, steals and other data. The data are analyzed and stored as csv format files by clustering algorithm.

摘 要

Abstract

摘要