本書是一本機器學習實用指南，提供從基礎知識到進階技能的全面學習路徑。本書以淺顯 易懂的方式介紹了機器學習的基本概念和主要類型，并詳細介紹使用 Python 及常見的庫進行數 據處理和機器學習的實操。此外，介紹了數據預處理的詳細過程，最后通過若干典型案例加深 讀者對機器學習的理解。本書適合對機器學習感興趣的初學者，也可作為軟件開發人員、數據分析師、學術研究人員的參考書籍。
謝雪葵，畢業于北郵軟件學院計算機科學系軟件工程專業。在校期間，多次獲得專業一、二等獎學金，并成功帶領團隊進行了校園APP的研發工作。阿誠網絡的創始人，該公司專注于為企業提供大數據相關服務。主要業務包括為企業提供大數據技術支持和降低成本、提高效率的解決方案，同時也提供基于機器學習的預測模型和智能決策支持。在過去的多年里，積累了豐富的企業級大數據項目實戰經驗，并負責大型銀行和互聯網公司的大數據項目開發和性能優化工作，其中包括使用機器學習技術進行風險評估、用戶行為分析和產品推薦等。
目錄
機器學習入門············································································1
機器學習簡介 ···········································································1
1.1.1 什么是機器學習································································1
1.1.2 機器學習的前景································································2
機器學習的主要類型 ··································································3
1.2.1 監督學習·········································································4
1.2.2 無監督學習······································································5
1.2.3 半監督學習······································································7
1.2.4 強化學習·········································································8
1.2.5 監督學習案例································································.10
選擇正確的算法·····································································.12
機器學習工具和環境·································································14
Python 介紹···········································································.14
2.1.1 Python 的安裝 ·······························································.14
2.1.2 Python 基礎語法 ····························································.19
2.1.3 Python 其他特性 ····························································.24
2.1.4 Python 簡單實戰案例(猜字游戲) ····································.31
2.1.5 Python 高級實戰案例(網絡爬蟲) ····································.35
數據科學庫···········································································.38
2.2.1 NumPy ········································································.38
2.2.2 Pandas ·········································································.45
2.2.3 數據科學庫案例(電商網站) ··········································.54
機器學習庫···········································································.55
2.3.1 Scikit-Learn···································································.55
2.3.2 TensorFlow ···································································.60
2.3.3 Keras···········································································.64
2.3.4 機器學習庫案例(預測糖尿病) ·······································.67
數據預處理·············································································70
數據導入 ··············································································.70
數據清洗 ··············································································.71
特征工程 ··············································································.73
3.3.1 特征選擇······································································.73
3.3.2 特征轉換······································································.75
3.3.3 特征縮放······································································.77
數據分割 ··············································································.78
3.4.1 訓練集·········································································.78
3.4.2 測試集·········································································.79
3.4.3 驗證集·········································································.80
案例分析:銀行客戶數據·························································.80
機器學習模型的構建與評估························································84
監督學習實戰········································································.84
4.1.1 線性回歸······································································.84
4.1.2 邏輯回歸······································································.86
4.1.3 決策樹·········································································.88
4.1.4 隨機森林······································································.90
無監督學習實戰·····································································.91
4.2.1 K-means ·······································································.92
4.2.2 主成分分析···································································.93
深度學習實戰········································································.95
4.3.1 神經網絡······································································.95
4.3.2 卷積神經網絡································································.98
4.3.3 循環神經網絡································································102
模型評估與選擇 ·····································································105
案例分析:客戶流失預測 ·························································107
第5章 5.1
機器學習項目實戰···································································111
項目一:房價預測 ·································································.111
5.1.1 數據獲取與理解·····························································112
5.1.2 數據預處理···································································116
5.1.3 特征工程······································································120
5.1.4 模型構建與訓練·····························································123
5.1.5 模型評估與優化·····························································125
5.1.6 結果解釋······································································128
項目二:圖像識別 ··································································130
5.2.1 數據獲取與理解·····························································131
5.2.2 數據預處理···································································134
5.2.3 特征工程······································································136
5.2.4 模型構建與訓練·····························································138
5.2.5 模型評估與優化·····························································140
5.2.6 結果解釋······································································143
項目三:自然語言處理 ····························································144
5.3.1 數據獲取與理解·····························································144
5.3.2 數據預處理···································································147
5.3.3 特征工程······································································148
5.3.4 模型構建與訓練·····························································149
5.3.5 模型評估與優化·····························································151
5.3.6 結果解釋······································································157
項目四:新聞主題分類 ····························································157
5.4.1 數據獲取與理解·····························································158
5.4.2 數據預處理···································································161
5.4.3 特征工程······································································164
5.4.4 模型構建與訓練·····························································166
5.4.5 模型評估與優化·····························································168
5.4.6 結果解釋······································································171
項目五:信用卡欺詐檢測 ·························································172
5.5.1 數據獲取與理解·····························································173
5.5.2 數據預處理···································································176
第5章 5.1
機器學習項目實戰···································································111
項目一:房價預測 ·································································.111
5.1.1 數據獲取與理解·····························································112
5.1.2 數據預處理···································································116
5.1.3 特征工程······································································120
5.1.4 模型構建與訓練·····························································123
5.1.5 模型評估與優化·····························································125
5.1.6 結果解釋······································································128
項目二:圖像識別 ··································································130
5.2.1 數據獲取與理解·····························································131
5.2.2 數據預處理···································································134
5.2.3 特征工程······································································136
5.2.4 模型構建與訓練·····························································138
5.2.5 模型評估與優化·····························································140
5.2.6 結果解釋······································································143
項目三:自然語言處理 ····························································144
5.3.1 數據獲取與理解·····························································144
5.3.2 數據預處理···································································147
5.3.3 特征工程······································································148
5.3.4 模型構建與訓練·····························································149
5.3.5 模型評估與優化·····························································151
5.3.6 結果解釋······································································157
項目四:新聞主題分類 ····························································157
5.4.1 數據獲取與理解·····························································158
5.4.2 數據預處理···································································161
5.4.3 特征工程······································································164
5.4.4 模型構建與訓練·····························································166
5.4.5 模型評估與優化·····························································168
5.4.6 結果解釋······································································171
項目五:信用卡欺詐檢測 ·························································172
5.5.1 數據獲取與理解·····························································173
5.5.2 數據預處理···································································176
新書資訊