[閒聊] 有趣的研究,<網>球-NBA的網絡結構

看板PHX-Suns (鳳凰城太陽)作者johneric (感染源)時間15年前 (2011/02/13 09:31)推噓9(9推 0噓 10→)

留言19則, 12人參與討論串1/1

來自這個活動 http://ppt.cc/GoGR 活動日期:4/1 Clippers @ Suns 活動內容: 1.打折的入場票 2.賽前講座,"Netball" 3.可以早點入場，在最前排看賽前練習 4.跟Suns舞者合照　　　　 5.贈品-復刻版的太陽球衣袋 6.特殊歡迎儀式 7.免費的輕軌車票研究大綱 http://ppt.cc/dilE 概要: : We are analyzing the network properties of basketball games. Defining : players as nodes and ball movements as links we can characterize a ball : game as samples of random networks. We have studied the playoff games : of the 2009/2010 NBA series. Using network theoretical constructs we : can determine average properties of these networks and characterize the : game, specific teams and specific players by their network properties. We : show that there is a typical network structure for an NBA team. Most : teams present a network that is close to this typical structure with the : Lakers (high clustering case) and the Jazz (low clustering case) marking : the extremes. The networks representing the Mavericks and the Cavaliers, : respectively, are very untypical. In addition, connectedness in the network : and the notion of flow centrality allows us to determine the value of a : player for the team, in contrast to the usual individualized statistics. : This is joint work with John Ingraham, and Alex Petersen from the : School of Mathematical and Statistical Sciences and with Jennifer Fewell : from the School of Life Sciences, Arizona State University. 我們著眼於分析籃球比賽的網絡特質。為此我們對球員和球的流動下定義，把場上球員當作節點，球的流動當做各節點之間的連結線；這樣每一場球賽都可以當做是隨機網絡裡的樣本。我們研究了２００９和２０１０的季後賽。利用網絡理論學說，我們可以決定這些網絡的平均特質；然後再利用這些特質，來歸納每個球隊和球員具有的網絡特性。我們證明出每個球隊都有一個典型的網絡結構。大部份球隊的網絡呈現是介於湖人（最高度群集網絡）和爵士（最低度群集網絡）之間　|　譯：　湖人<-----------------其他隊--------------------->爵士 | | 我猜是因為湖人三角進攻戰術的關係? 才會有最高度群集的現像 | 　|　　而爵士的戰術導致球在各球員間的流動差不多? | 但有二隊網絡呈現很不符合典型現像，就是小牛和騎士隊。 |　譯：騎士我猜是大部份球是集中在LBJ身上，而小牛則猜不出來　　　| 除此之外，網絡裡的連通性和球流動的集中性，可以讓我們判斷該球員對球隊的價值；這可以用來和傳統的個人化數據比較一下。這份研究是由二位ＡＳＵ數學和統計系教授和一位人文科學系的教授所完成。如果想看最終研究成果，請參加4/1的活動。 -- ※ 發信站: 批踢踢實業坊(ptt.cc) ◆ From: 98.165.229.30 ※ 編輯: johneric 來自: 98.165.229.30 (02/13 09:32)

推

abcd810112

02/13 09:40, , 1^F

02/13 09:40, 1^F

→

henryokc

02/13 09:51, , 2^F

02/13 09:51, 2^F

推

z1233218

02/13 09:58, , 3^F

02/13 09:58, 3^F