交通時空大數據分析、挖掘與可視化（Python版）

大數據時代已經到來，隨著數據的逐步開放，交通領域的研究課題或多或少都要接觸、使用時空 大數據。交通領域的從業者迫切需要強有力的工具和技術應對日益紛雜的交通數據。交通是一個交叉 學科，交通數據分析人才的知識體系需要與數據處理、網絡爬蟲、數據可視化、地理信息、復雜網絡、 數據挖掘、機器學習等多學科知識深度融合，這也為交通領域的人才培養帶來巨大挑戰。 在此背景下，本書針對不同的學習階段與業務需求設計了三篇共15章內容。基礎篇（第1～5章） 梳理Python數據分析、網絡爬蟲、數據可視化、地理信息等基礎知識；應用篇（第6～10章）介紹 出租車GPS數據、地鐵IC刷卡數據、共享單車訂單數據、公交GPS數據等各類時空大數據的實際案 例應用；方法篇（第11～15章）融匯數據挖掘、空間統計、復雜網絡學科等交叉學科方法，與交通 領域的大量實際案例分析結合，全面梳理總結交通時空大數據所需跨學科技能。 本書由淺入深，學科交叉，強調實踐。對讀者不同的學習階段與業務需求設計相應內容，全面梳 理總結交通大數據科研所需技能，並與交通領域的大量實際案例分析結合。本書可作為教材也可作為 參考工具書，基礎篇定位交通數據領域新手入門，應用篇定位有數據分析需求的高校學生或社會人士， 方法篇定位高校學術科研人員。

目　錄

基 礎 篇

第1章　緒論 ·····························2

1.1 多源交通時空大數據簡介 ················2

1.1.1 傳統集計統計數據 ·······························3

1.1.2 個體連續追蹤數據 ·······························4

1.1.3 地理空間信息數據 ·······························5

1.2 為什麽要用Python處理交通大數據 ·····6

1.2.1 常用數據處理技術 ·······························6

1.2.2 Python在交通大數據領域中的優勢 ····8

1.2.3 Python與SQL的比較 ····························9

1.3 大規模數據處理的解決方案··············9

1.3.1 決定大數據處理性能的三個硬件

 要素 ·······················································9

1.3.2 分佈式數據處理架構 ·························11

1.4 本章習題 ···································14

第2章　Python數據處理基礎 ······15

2.1 Python的環境配置 ························15

2.1.1 Python的集成開發環境 ······················15

2.1.2 Anaconda的安裝 ·································16

2.1.3 Jupyter Notebook的使用 ·····················16

2.1.4 Python第三方庫的安裝 ······················18

2.2 Python基本語法 ···························19

2.2.1 對象與變量 ·········································19

2.2.2 運算符 ·················································20

2.2.3 內置數據類型 ·····································20

2.2.4 語句 ·····················································24

2.2.5 函數 ·····················································26

2.2.6 包的使用 ·············································27

2.2.7 數據分析常用第三方庫簡介 ·············28

2.3 pandas數據處理基礎 ·····················29

2.3.1 數據文件的編碼格式與存儲形式 ·····30

2.3.2 數據表的行列處理 ·····························33

2.3.3 數據的表格運算 ·································41

2.4 時空大數據的處理思維 ·················46

2.4.1 復雜數據處理任務的解決思路 ·········46

2.4.2 數據處理任務分解實例：地鐵換乘量

 識別 ······················································49

2.5 數據處理中表格運算的常用技巧 ······51

2.5.1 分組編號 ·············································51

2.5.2 去除重復的記錄 ·································53

2.5.3 個體ID重新編號 ·································54

2.5.4 生成數據之間的對應表 ·····················55

2.5.5 時空插值 ·············································58

2.6 本章習題 ···································60

2.6.1 思考題 ·················································60

2.6.2 Python基礎代碼練習 ··························60

2.6.3 pandas基礎代碼練習 ··························62

第3章　數據可視化基礎 ············64

3.1 可視化的基本原則 ·······················64